From manoj at phys.ufl.edu Fri May 1 07:50:56 2009 From: manoj at phys.ufl.edu (Manoj Srivastava) Date: Fri, 1 May 2009 01:50:56 -0400 (EDT) Subject: [Pw_forum] transmission calculation In-Reply-To: <1241096915.5731.36.camel@alex-laptop> Message-ID: Dear Alexander, Thank you for your response. On Thu, 30 Apr 2009, Alexander Smogunov wrote: > Dear Manoj > Please provide also scf files for the leads. > Note that in the case ikind=2 (different leads) the scattering file > should be a double unit cell i.e. should contain the scattering barrier > + some portion of the right lead + the scattering barrier in order to You meant scatterin barrier+some portion of right lead+some portion of left lead, right? > have regular periodicity in the z direction. You should also provide > bds parameter which says where the scattering barrier ends (while it > should start from 0). > Regards, Alexander > > > On Thu, 2009-04-30 at 07:23 -0400, Manoj Srivastava wrote: > > Dear PWSCF users and developers, > > I have tried to do a transmission calcuation and not getting right > > results. I have tried to figure out the reason behind this and realized > > that in my set up for some reason the Bloch waves in left lead get changed > > resulting wrong transmission. By this I meant that, if I just try to > > calculate complex band of the left lead, I have a kz for each kx,ky and E > > but when I am doing transmission calculation, the Bloch's wave in the same > > lead has different kz for the same kx,ky and E, which does not make sense > > as why should the Bloch's wave in lead get affected by the presence of > > scattering region? For the right lead kz remains the same in both cases > > though, which I believe how it should be. > > I did this for several cases, I interchanged right and left leads, > > increased number of atoms in the scattering region, chose scattering > > region in different way, and everytime kz of left lead changes. Following > > is my input file for the scattering region - > > > > scf part - > > > > &control > > calculation='scf' > > pseudo_dir = '/home/manoj/espresso-4.0.1/pseudo', > > outdir='./', > > prefix='cus', > > / > > &system > > ibrav =0, > > celldm(1)=4.82, > > nat= 10, > > ntyp= 1, > > ecutwfc =50.0, > > occupations='smearing', > > smearing='gaussian', > > degauss=0.02, > > ecutrho=500 > > / > > &electrons > > conv_thr = 1.0e-8 > > mixing_beta=0.7 > > / > > ATOMIC_SPECIES > > Cu 63.55 Cu.pz-d-rrkjus.UPF > > ATOMIC_POSITIONS > > Cu 0.0 0.0 0.0 > > Cu -0.5 0.288675134 0.816496581 > > Cu -1.0 0.577350269 1.632993162 > > Cu -1.5 0.866025404 2.449489743 > > Cu 0.0 0.0 3.265986324 > > Cu 0.5 -0.288675134 4.082482905 > > Cu 1.0 -0.577350269 4.898979486 > > Cu 1.5 -0.866025404 5.715476066 > > Cu 1.0 -0.577350269 6.531972647 > > Cu 0.5 -0.288675134 7.348469228 > > K_POINTS (automatic) > > 6 6 4 1 1 1 > > CELL_PARAMETERS {hexagonal} > > 1.0 0.0 0.0 > > -0.5 0.866025403 0.0 > > 0.0 0.0 8.16496581 > > > > > > > > conductance part - > > > > &inputcond > > outdir='./' > > prefixl='cul' > > prefixr='cur' > > prefixs='cus' > > tran_file ='trans.twin' > > ikind=2 > > ecut2d=50 > > energy0=0.0 > > denergy=0.0 > > ewind=5.d0 > > epsproj=1.d-6 > > nz1=25 > > / > > 2 > > 0.000000 0.000000 1 > > 0.5000000 -0.5000000 1 > > > > 1 > > 0.0 > > > > Thank you for your attention. > > > > Regards, > > Manoj Srivastava > > Graduate Student > > Department of Physics > > University of Florida, Gainesville, FL > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > -- > Alexander Smogunov > Email: smogunov at sissa.it > Home page: http://people.sissa.it/~smogunov > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From jasius_1 at yahoo.com Fri May 1 16:04:39 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Fri, 1 May 2009 07:04:39 -0700 (PDT) Subject: [Pw_forum] Conjugate Gradient minimization not converged after 50 iterations In-Reply-To: <3CCBD25C-35E9-468D-8AD0-0085605460F6@democritos.it> Message-ID: <977003.29732.qm@web54402.mail.yahoo.com> Dear all, while running normal mode calculation on a large system I can see that conjugate gradient minimization does not converge after 50 iterations in some modes. By inspection, some modes converge in 46-48 iterations. Would increasing number of iterations solve that problem? Is it a problem at all as residual after 50 iterations is quite small, does that mean it's acceptable as calculations continue? *** Starting Conjugate Gradient minimization *** d2ion: alpha = 0.50 *** mode # 1 : using asr *** mode # 2 : using asr *** mode # 3 : using asr *** mode # 4 : 46 iterations *** mode # 5 : 45 iterations *** mode # 6 : 48 iterations *** Conjugate Gradient minimization not converged after 50 iterations residual norm |Ax-b|^2 : 0.1718E-13 *** mode # 7 : 50 iterations *** Conjugate Gradient minimization not converged after 50 iterations residual norm |Ax-b|^2 : 0.1148E-13 *** mode # 8 : 50 iterations thanks Jonas Baltrusaitis University of Iowa From kvvamsi20 at gmail.com Sat May 2 12:25:06 2009 From: kvvamsi20 at gmail.com (K V Vamsi) Date: Sat, 2 May 2009 15:55:06 +0530 Subject: [Pw_forum] charge density units? Message-ID: Hi, I am vamsi from IISc bangalore.............. I am using the package for quite some time ........... I have plotted the charge density 2D plots.......... wat are the unitsexactly ? there is a minimum and maximum value of thegrid what does it mean exactly.........? Do this calculations include total charge or the valence charge? please let me the answers for these questions...................... -- Regards K V Vamsi Materials Engineering Deformation mechanisms and modelling group IISc, Bangalore - 560 012 Mob : 09886818221 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090502/0d5b77f8/attachment.htm From paulatto at sissa.it Sat May 2 13:27:00 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 2 May 2009 13:27:00 +0200 (CEST) Subject: [Pw_forum] charge density units? In-Reply-To: References: Message-ID: <12642.82.48.143.248.1241263620.squirrel@webmail.sissa.it> On Sat, May 2, 2009 12:25, K V Vamsi wrote: > Hi, > I am vamsi from IISc bangalore.............. > I am using the package for quite some time ........... > I have plotted the charge density 2D plots.......... wat are the > unitsexactly ? > there is a minimum and maximum value of thegrid what does it mean > exactly.........? please have a look at the mailing list archives, where the question has been asked and answered several times: > Do this calculations include total charge or the valence charge? > please let me the answers for these questions...................... only valence charge is included, if you are using ultrasoft pseudopotential augmentation charge should be included as well. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From jasius_1 at yahoo.com Sun May 3 06:00:05 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Sat, 2 May 2009 21:00:05 -0700 (PDT) Subject: [Pw_forum] DOS virtual band calculations Message-ID: <389832.24381.qm@web54410.mail.yahoo.com> Hi, in DOS calculations manual I couldn't find anything mentioned related on empty band calculations. So my question is: how do I add virtual (empty) band DOS calculation or is it done by default? Is there a way to specify empty bands included in calculations? I would like to have both occupied and virtual DOS calculated thanks Jonas Baltrusaitis University of Iowa From degironc at sissa.it Sun May 3 10:24:28 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sun, 03 May 2009 10:24:28 +0200 Subject: [Pw_forum] DOS virtual band calculations In-Reply-To: <389832.24381.qm@web54410.mail.yahoo.com> References: <389832.24381.qm@web54410.mail.yahoo.com> Message-ID: <49FD54BC.7000001@sissa.it> You can use nbnd in pw.x input to define the number of bands to be calculated. Stefano de Gironcoli - SISSA and DEMOCRITOS Jonas Baltrusaitis wrote: > Hi, > > in DOS calculations manual I couldn't find anything mentioned related on empty band calculations. So my question is: how do I add virtual (empty) band DOS calculation or is it done by default? Is there a way to specify empty bands included in calculations? I would like to have both occupied and virtual DOS calculated > thanks > Jonas Baltrusaitis > University of Iowa > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From vitruss at gmail.com Sun May 3 14:36:45 2009 From: vitruss at gmail.com (Vit) Date: Sun, 3 May 2009 19:36:45 +0700 Subject: [Pw_forum] Compilation issue (phenom II and ACML) Message-ID: <11f280750905030536r1b52bd83x1df4a5d35ee3731d@mail.gmail.com> Good day, everyone! I'm trying to use PWSCF on my own computer with Phenom II x4 with Arch Linux onboard. The User guide says that I can get huge benefit from using ACML from AMD. I've downloaded and installed it. Now I'm trying to compile QE with this library. If I use just "configure" it uses internal BLAS, LAPACK and FFTW libraries. If I try LDFLAGS="-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml" it says: checking build system type... x86_64-unknown-linux-gnu checking architecture... amd64 checking for Fortran 77 compiler default output file name... configure: error: Fortran 77 compiler cannot create executables See `config.log' for more details. What's wrong? Config.log: This file contains any messages produced by compilers while running configure, to aid debugging if configure makes a mistake. It was created by ESPRESSO configure 4.0, which was generated by GNU Autoconf 2.59. Invocation command line was $ ./configure F90=mpif90 F77=mpif77 LDFLAGS=-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml ## --------- ## ## Platform. ## ## --------- ## hostname = Arch uname -m = x86_64 uname -r = 2.6.29-ARCH uname -s = Linux uname -v = #1 SMP PREEMPT Wed Apr 29 15:36:46 CEST 2009 /usr/bin/uname -p = unknown /bin/uname -X = unknown /bin/arch = x86_64 /usr/bin/arch -k = unknown /usr/convex/getsysinfo = unknown hostinfo = unknown /bin/machine = unknown /usr/bin/oslevel = unknown /bin/universe = unknown PATH: /opt/wine/bin PATH: /bin PATH: /usr/bin PATH: /sbin PATH: /usr/sbin PATH: /usr/bin/perlbin/site PATH: /usr/bin/perlbin/vendor PATH: /usr/bin/perlbin/core ## ----------- ## ## Core tests. ## ## ----------- ## configure:1398: checking build system type configure:1416: result: x86_64-unknown-linux-gnu configure:1665: checking for Fortran 77 compiler version configure:1668: mpif77 --version &5 GNU Fortran (GCC) 4.4.0 Copyright (C) 2009 Free Software Foundation, Inc. GNU Fortran comes with NO WARRANTY, to the extent permitted by law. You may redistribute copies of GNU Fortran under the terms of the GNU General Public License. For more information about these matters, see the file named COPYING configure:1671: $? = 0 configure:1673: mpif77 -v &5 Using built-in specs. Target: x86_64-unknown-linux-gnu Configured with: ../configure --prefix=/usr --enable-shared --enable-languages=c,c++,fortran,objc,obj-c++ --enable-threads=posix --mandir=/usr/share/man --infodir=/usr/share/info --enable-__cxa_atexit --disable-multilib --libdir=/usr/lib --libexecdir=/usr/lib --enable-clocale=gnu --disable-libstdcxx-pch --with-tune=generic Thread model: posix gcc version 4.4.0 (GCC) configure:1676: $? = 0 configure:1678: mpif77 -V &5 gfortran: '-V' option must have argument configure:1681: $? = 1 configure:1695: checking for Fortran 77 compiler default output file name configure:1698: mpif77 -L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml conftest.f >&5 /usr/bin/ld: cannot find -lacml collect2: ld returned 1 exit status configure:1701: $? = 1 configure: failed program was: | program main | | end configure:1740: error: Fortran 77 compiler cannot create executables See `config.log' for more details. ## ---------------- ## ## Cache variables. ## ## ---------------- ## ac_cv_build=x86_64-unknown-linux-gnu ac_cv_build_alias=x86_64-unknown-linux-gnu ac_cv_env_CC_set= ac_cv_env_CC_value= ac_cv_env_CFLAGS_set= ac_cv_env_CFLAGS_value= ac_cv_env_CPPFLAGS_set= ac_cv_env_CPPFLAGS_value= ac_cv_env_CPP_set= ac_cv_env_CPP_value= ac_cv_env_F77_set=set ac_cv_env_F77_value=mpif77 ac_cv_env_FFLAGS_set= ac_cv_env_FFLAGS_value= ac_cv_env_LDFLAGS_set=set ac_cv_env_LDFLAGS_value='-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml' ac_cv_env_build_alias_set= ac_cv_env_build_alias_value= ac_cv_env_host_alias_set= ac_cv_env_host_alias_value= ac_cv_env_target_alias_set= ac_cv_env_target_alias_value= ## ----------------- ## ## Output variables. ## ## ----------------- ## CC='' CFLAGS='' CPP='' CPPFLAGS='' DEFS='' ECHO_C='' ECHO_N='-n' ECHO_T='' EGREP='' EXEEXT='' F77='mpif77' FFLAGS='' FLIBS='' LDFLAGS='-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml' LIBOBJS='' LIBS='' LTLIBOBJS='' OBJEXT='' PACKAGE_BUGREPORT='' PACKAGE_NAME='ESPRESSO' PACKAGE_STRING='ESPRESSO 4.0' PACKAGE_TARNAME='espresso' PACKAGE_VERSION='4.0' PATH_SEPARATOR=':' SET_MAKE='' SHELL='/bin/sh' SIZEOF_INT_P='' ac_ct_CC='' ac_ct_F77='' ar='' arflags='' arflags_dynamic='' bindir='${exec_prefix}/bin' blas_libs='' blas_line='' build='x86_64-unknown-linux-gnu' build_alias='' build_cpu='x86_64' build_os='linux-gnu' build_vendor='unknown' cc='' cflags='' cpp='' cppflags='' datadir='${prefix}/share' dflags='' exec_prefix='NONE' f77='mpif77' f90='mpif90' f90flags='' f90rule='' fdflags='' fflags='' fflags_noopt='' fft_libs='' fft_line='' host_alias='' iflags='' imod='' includedir='${prefix}/include' includefftw='' infodir='${prefix}/info' lapack_libs='' lapack_line='' ld='' ldflags='-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml' libdir='${exec_prefix}/lib' libexecdir='${exec_prefix}/libexec' localstatedir='${prefix}/var' mandir='${prefix}/man' mass_libs='' mass_line='' mpi_libs='' mpi_line='' mpif90='' oldincludedir='/usr/include' parallel_report='' pgplot_libs='' pre_fdflags='' prefix='NONE' program_transform_name='s,x,x,' ranlib='' sbindir='${exec_prefix}/sbin' sharedstatedir='${prefix}/com' sysconfdir='${prefix}/etc' target_alias='' ## ----------- ## ## confdefs.h. ## ## ----------- ## #define PACKAGE_BUGREPORT "" #define PACKAGE_NAME "ESPRESSO" #define PACKAGE_STRING "ESPRESSO 4.0" #define PACKAGE_TARNAME "espresso" #define PACKAGE_VERSION "4.0" configure: exit 77 From jasius_1 at yahoo.com Sun May 3 15:42:18 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Sun, 3 May 2009 06:42:18 -0700 (PDT) Subject: [Pw_forum] DOS virtual band calculations In-Reply-To: <49FD54BC.7000001@sissa.it> Message-ID: <928590.23288.qm@web54407.mail.yahoo.com> Thanks, but how do I define virtuals? Nbnd defines number of total bands jonas --- On Sun, 5/3/09, Stefano de Gironcoli wrote: > From: Stefano de Gironcoli > Subject: Re: [Pw_forum] DOS virtual band calculations > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Sunday, May 3, 2009, 1:24 AM > You can use nbnd in pw.x input to define the number of bands > to be calculated. > Stefano de Gironcoli - SISSA and DEMOCRITOS > > > Jonas Baltrusaitis wrote: > > Hi, > > > > in DOS calculations manual I couldn't find > anything mentioned related on empty band calculations. So my > question is: how do I add virtual (empty) band DOS > calculation or is it done by default? Is there a way to > specify empty bands included in calculations? I would like > to have both occupied and virtual DOS calculated > > thanks > > Jonas Baltrusaitis > > University of Iowa > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > From somesh.kb at gmail.com Sun May 3 15:58:12 2009 From: somesh.kb at gmail.com (Somesh Kumar Bhattacharya) Date: Sun, 3 May 2009 15:58:12 +0200 Subject: [Pw_forum] DOS virtual band calculations In-Reply-To: <928590.23288.qm@web54407.mail.yahoo.com> References: <49FD54BC.7000001@sissa.it> <928590.23288.qm@web54407.mail.yahoo.com> Message-ID: Hi Jonas, nbnd defines both occipied and unoccupied bands. Ciao, On Sun, May 3, 2009 at 3:42 PM, Jonas Baltrusaitis wrote: > > Thanks, but how do I define virtuals? Nbnd defines number of total bands > > jonas > > > --- On Sun, 5/3/09, Stefano de Gironcoli wrote: > > > From: Stefano de Gironcoli > > Subject: Re: [Pw_forum] DOS virtual band calculations > > To: jasius_1 at yahoo.com, "PWSCF Forum" > > Date: Sunday, May 3, 2009, 1:24 AM > > You can use nbnd in pw.x input to define the number of bands > > to be calculated. > > Stefano de Gironcoli - SISSA and DEMOCRITOS > > > > > > Jonas Baltrusaitis wrote: > > > Hi, > > > > > > in DOS calculations manual I couldn't find > > anything mentioned related on empty band calculations. So my > > question is: how do I add virtual (empty) band DOS > > calculation or is it done by default? Is there a way to > > specify empty bands included in calculations? I would like > > to have both occupied and virtual DOS calculated > > > thanks > > > Jonas Baltrusaitis > > > University of Iowa > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Somesh Kr. Bhattacharya Post Doctoral Fellow Room No. 263, Leonardo Building, The Abdus Salam International Centre for Theoretical Physics Strada Costiera, 11 I-34014 Trieste Italy Phone: +39-040-2240399 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090503/f8c6dd88/attachment.htm From baroni at sissa.it Sun May 3 17:31:28 2009 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 3 May 2009 17:31:28 +0200 Subject: [Pw_forum] DOS virtual band calculations In-Reply-To: <928590.23288.qm@web54407.mail.yahoo.com> References: <928590.23288.qm@web54407.mail.yahoo.com> Message-ID: number of virtuals = number of bands (nbnd) - number of occupied states SB On May 3, 2009, at 3:42 PM, Jonas Baltrusaitis wrote: > > Thanks, but how do I define virtuals? Nbnd defines number of total > bands > > jonas > > > --- On Sun, 5/3/09, Stefano de Gironcoli wrote: > >> From: Stefano de Gironcoli >> Subject: Re: [Pw_forum] DOS virtual band calculations >> To: jasius_1 at yahoo.com, "PWSCF Forum" >> Date: Sunday, May 3, 2009, 1:24 AM >> You can use nbnd in pw.x input to define the number of bands >> to be calculated. >> Stefano de Gironcoli - SISSA and DEMOCRITOS >> >> >> Jonas Baltrusaitis wrote: >>> Hi, >>> >>> in DOS calculations manual I couldn't find >> anything mentioned related on empty band calculations. So my >> question is: how do I add virtual (empty) band DOS >> calculation or is it done by default? Is there a way to >> specify empty bands included in calculations? I would like >> to have both occupied and virtual DOS calculated >>> thanks >>> Jonas Baltrusaitis >>> University of Iowa >>> >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From yaldachem at gmail.com Sun May 3 21:23:49 2009 From: yaldachem at gmail.com (yaldaa kh) Date: Sun, 3 May 2009 12:23:49 -0700 Subject: [Pw_forum] dos Message-ID: <47431d9f0905031223v57394840ye5320f91c9da05ce@mail.gmail.com> dear all user How can I project dos to molecular orbital? what is the importance of nbnd in calculations for dos? and for plottingdos or pdos, if the fermi energy is assumed to be zero automatically by program? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090503/924ca4f2/attachment.htm From yaldachem at gmail.com Sun May 3 21:28:31 2009 From: yaldachem at gmail.com (yaldaa kh) Date: Sun, 3 May 2009 12:28:31 -0700 Subject: [Pw_forum] dos In-Reply-To: <47431d9f0905031223v57394840ye5320f91c9da05ce@mail.gmail.com> References: <47431d9f0905031223v57394840ye5320f91c9da05ce@mail.gmail.com> Message-ID: <47431d9f0905031228k2119d5b6x5f56bf55589ee9a2@mail.gmail.com> ---------- Forwarded message ---------- From: yaldaa kh Date: May 3, 2009 12:23 PM Subject: dos To: pw_forum dear all user How can I project dos to molecular orbital? what is the importance of nbnd in calculations for dos? and for plottingdos or pdos, if the fermi energy is assumed to be zero automatically by program? thanks yalda , -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090503/ddae30fb/attachment.htm From prasenjit.jnc at gmail.com Sun May 3 23:00:29 2009 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Sun, 3 May 2009 23:00:29 +0200 Subject: [Pw_forum] dos In-Reply-To: <47431d9f0905031223v57394840ye5320f91c9da05ce@mail.gmail.com> References: <47431d9f0905031223v57394840ye5320f91c9da05ce@mail.gmail.com> Message-ID: <627e0ffa0905031400l3cf1525bsfb7b22f789195085@mail.gmail.com> dear all user > Dear Yaldaa, > How can I project dos to molecular orbital? > There is a post-processing tool called projwfc.x which you can use to project dos to molecular orbital. what is the importance of nbnd in calculations for dos? > e "nbnd" tells the code the no. of Kohn-Sham states it has to compute.....While computing the dos, one wants to plot the dos for both the occupied & unoccupied manifold, the no. of energy levels is read in by the dos.x from the output of your scf/nscf run. So while doing an scf/ nscf calculations you can control the total no. of states with the variable "nbnd". Depending on the type of occupation scheme you choose the value of nbnd is nelec/2 (fixed occupation) or nelec/2+some extra bands (when you use smearing). If you want to plot the dos, usually you should add some extra levels using nbnd & do the scf or nscf calculations to get the unoccupied manifold. and for plottingdos or pdos, if the fermi energy is assumed to be zero > automatically by program? > No, you have to do the alignment yourself With regards, Prasenjit. -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090503/c792a8d6/attachment-0001.htm From jasius_1 at yahoo.com Sun May 3 23:59:34 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Sun, 3 May 2009 14:59:34 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <627e0ffa0905031400l3cf1525bsfb7b22f789195085@mail.gmail.com> Message-ID: <770442.92952.qm@web54403.mail.yahoo.com> Prasenjit, could you explain why both scf followed by nscf calculations are needed for DOS calculations, as per example 8? If only scf density is needed for further projection, why not to calculate only one? Jonas --- On Sun, 5/3/09, Prasenjit Ghosh wrote: > From: Prasenjit Ghosh > Subject: Re: [Pw_forum] dos > To: "PWSCF Forum" > Date: Sunday, May 3, 2009, 2:00 PM > dear all user > > > > Dear Yaldaa, > > > > How can I project dos to molecular orbital? > > > > There is a post-processing tool called projwfc.x which you > can use to > project dos to molecular orbital. > > what is the importance of nbnd in calculations for dos? > > > e > "nbnd" tells the code the no. of Kohn-Sham > states it has to > compute.....While computing the dos, one wants to plot the > dos for both the > occupied & unoccupied manifold, the no. of energy > levels is read in by the > dos.x from the output of your scf/nscf run. So while doing > an scf/ nscf > calculations you can control the total no. of states with > the variable > "nbnd". > Depending on the type of occupation scheme you choose the > value of nbnd is > nelec/2 (fixed occupation) or nelec/2+some extra bands > (when you use > smearing). If you want to plot the dos, usually you should > add some extra > levels using nbnd & do the scf or nscf calculations to > get the unoccupied > manifold. > > and for plottingdos or pdos, if the fermi energy is assumed > to be zero > > automatically by program? > > > > No, you have to do the alignment yourself > > With regards, > > Prasenjit. > > > > -- > PRASENJIT GHOSH, > POST-DOC, > ROOM NO: 265, MAIN BUILDING, > CM SECTION, ICTP, > STRADA COSTERIA 11, > TRIESTE, 34104, > ITALY > PHONE: +39 040 2240 369 (O) > +39 3807528672 (M) > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From akohlmey at cmm.chem.upenn.edu Mon May 4 01:19:53 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 3 May 2009 19:19:53 -0400 Subject: [Pw_forum] Compilation issue (phenom II and ACML) In-Reply-To: <11f280750905030536r1b52bd83x1df4a5d35ee3731d@mail.gmail.com> References: <11f280750905030536r1b52bd83x1df4a5d35ee3731d@mail.gmail.com> Message-ID: <7b6913e90905031619l507d8df7ta81a53d8ec2e819b@mail.gmail.com> On Sun, May 3, 2009 at 8:36 AM, Vit wrote: > Good day, everyone! > I'm trying to use PWSCF on my own computer with Phenom II x4 with Arch > Linux onboard. The User guide says that I can get huge benefit from > using ACML from AMD. > I've downloaded and installed it. Now I'm trying to compile QE with > this library. If I use just "configure" it uses internal BLAS, LAPACK > and FFTW libraries. If I try > LDFLAGS="-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml" it says: hmmm... if you want to run with MPI, you should not use the multi-threaded ACML but the serial one. see the comments on using OMP_NUM_THREADS=1 with intel's MKL v10. > checking build system type... x86_64-unknown-linux-gnu > checking architecture... amd64 > checking for Fortran 77 compiler default output file name... > configure: error: Fortran 77 compiler cannot create executables > See `config.log' for more details. > > What's wrong? [...] > configure:1695: checking for Fortran 77 compiler default output file > name > configure:1698: mpif77 ?-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml > conftest.f ?>&5 > /usr/bin/ld: cannot find -lacml > collect2: ld returned 1 exit status looks like your ACML installation does not work. did you test it on a simpler program before? cheers, axel > configure:1701: $? = 1 > configure: failed program was: > | ? ? ? program main > | > | ? ? ? end > configure:1740: error: Fortran 77 compiler cannot create executables > See `config.log' for more details. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From prasenjit.jnc at gmail.com Mon May 4 09:04:18 2009 From: prasenjit.jnc at gmail.com (Prasenjit Ghosh) Date: Mon, 4 May 2009 09:04:18 +0200 Subject: [Pw_forum] dos In-Reply-To: <770442.92952.qm@web54403.mail.yahoo.com> References: <627e0ffa0905031400l3cf1525bsfb7b22f789195085@mail.gmail.com> <770442.92952.qm@web54403.mail.yahoo.com> Message-ID: <627e0ffa0905040004j72f19d2fk7f8befa300ad06be@mail.gmail.com> Hi Jonas, For DOS calculations one needs to do an integration over the kpts in the Brillouin zone. So you need a very fine k-point mesh than necessary for a scf calculation. Moreover you need a large number of bands. Doing an scf calculation with such a fine k-point mesh and so many bands is very expensive. So you should first do an scf calculation with the min. no. of k-points and bands and then do a nscf calculation with a finer k-point and larger no. of bands. In the nscf step, the code does an interpolation for the extra bands and k-points and not a self-consistent calculation. This saves computational time. Hope this helps. Prasenjit. 2009/5/3 Jonas Baltrusaitis > > Prasenjit, > > could you explain why both scf followed by nscf calculations are needed for > DOS calculations, as per example 8? If only scf density is needed for > further projection, why not to calculate only one? > > Jonas > > > --- On Sun, 5/3/09, Prasenjit Ghosh wrote: > > > From: Prasenjit Ghosh > > Subject: Re: [Pw_forum] dos > > To: "PWSCF Forum" > > Date: Sunday, May 3, 2009, 2:00 PM > > dear all user > > > > > > > Dear Yaldaa, > > > > > > > How can I project dos to molecular orbital? > > > > > > > There is a post-processing tool called projwfc.x which you > > can use to > > project dos to molecular orbital. > > > > what is the importance of nbnd in calculations for dos? > > > > > e > > "nbnd" tells the code the no. of Kohn-Sham > > states it has to > > compute.....While computing the dos, one wants to plot the > > dos for both the > > occupied & unoccupied manifold, the no. of energy > > levels is read in by the > > dos.x from the output of your scf/nscf run. So while doing > > an scf/ nscf > > calculations you can control the total no. of states with > > the variable > > "nbnd". > > Depending on the type of occupation scheme you choose the > > value of nbnd is > > nelec/2 (fixed occupation) or nelec/2+some extra bands > > (when you use > > smearing). If you want to plot the dos, usually you should > > add some extra > > levels using nbnd & do the scf or nscf calculations to > > get the unoccupied > > manifold. > > > > and for plottingdos or pdos, if the fermi energy is assumed > > to be zero > > > automatically by program? > > > > > > > No, you have to do the alignment yourself > > > > With regards, > > > > Prasenjit. > > > > > > > > -- > > PRASENJIT GHOSH, > > POST-DOC, > > ROOM NO: 265, MAIN BUILDING, > > CM SECTION, ICTP, > > STRADA COSTERIA 11, > > TRIESTE, 34104, > > ITALY > > PHONE: +39 040 2240 369 (O) > > +39 3807528672 (M) > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- PRASENJIT GHOSH, POST-DOC, ROOM NO: 265, MAIN BUILDING, CM SECTION, ICTP, STRADA COSTERIA 11, TRIESTE, 34104, ITALY PHONE: +39 040 2240 369 (O) +39 3807528672 (M) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/a79fcacd/attachment.htm From giannozz at democritos.it Mon May 4 09:57:22 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 04 May 2009 09:57:22 +0200 Subject: [Pw_forum] phonon calculation for large system In-Reply-To: <3ebf88c60904301114q2f857669mdf79bd810bf208f7@mail.gmail.com> References: <3ebf88c60904291515g33a39565w68f289e6f75a021b@mail.gmail.com> <49F95761.5010404@sissa.it> <97BACDB3-4033-4612-B470-E9CB8E943E51@democritos.it> <49F95DBA.8000803@sissa.it> <3ebf88c60904301114q2f857669mdf79bd810bf208f7@mail.gmail.com> Message-ID: <49FE9FE2.5030001@democritos.it> zq wu wrote: > Can single-q, single-mode calculation run in several nodes yes P. -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at democritos.it Mon May 4 10:02:38 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 04 May 2009 10:02:38 +0200 Subject: [Pw_forum] Conjugate Gradient minimization not converged after 50 iterations In-Reply-To: <977003.29732.qm@web54402.mail.yahoo.com> References: <977003.29732.qm@web54402.mail.yahoo.com> Message-ID: <49FEA11E.70605@democritos.it> Jonas Baltrusaitis wrote: > conjugate gradient minimization does not converge after 50 iterations in some modes. > By inspection, some modes converge in 46-48 iterations. Would increasing number of > iterations solve that problem? likely, yes > Is it a problem at all as residual after 50 iterations is quite small maybe it isn't a problem at all. It simply means that the prefixed threshold on the residual vector has not been reached. How this translates into errors in the final results, it is hard to say. One should go to the definition of the threshold and try to figure out an estimate for the corresponding error on the frequeencies; or else, one should try a test casse with different thresholds Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From shruba at gmail.com Mon May 4 10:17:47 2009 From: shruba at gmail.com (shruba at gmail.com) Date: Mon, 4 May 2009 04:17:47 -0400 Subject: [Pw_forum] problem with installing Quantum ESPRESSO CVS version In-Reply-To: References: <822f4ec80904291443h155ecd7jf16c9fd0bb92e3b3@mail.gmail.com> Message-ID: <822f4ec80905040117t4efb9e80haa2d4662dbd1ab8f@mail.gmail.com> Dear Dr. Giannozzi, Thanks a lot for the reply, we successfully able to install it in our cluster after correcting the library locations in make system. Thanks again for your help. Regards Shruba Gangopadhyay On Thu, Apr 30, 2009 at 3:16 AM, Paolo Giannozzi wrote: > > On Apr 29, 2009, at 23:43 , shruba at gmail.com wrote: > > /apps/intel/fce/10.1.008/lib/libimf.so: warning: warning: feupdateenv is > not implemented > > and will always fail > > > not a problem > > /opt/intel/mkl/10.0.1.014/lib/em64t/libmkl_lapack.so: undefined reference > to `mkl_serv_load_fun' > > > in your make.sys, you should have > LDFLAGS = -i-static -openmp > BLAS_LIBS = -L /opt/intel/mkl/10.0.1.014/lib/em64t -lmkl_em64t > LAPACK_LIBS= > and no -lmkl_lapack -lguide -lpthreads > > when I contacted to the cluster management they suggested me to check > > weather CVS version is compatible with version 10.0.1.014 of the intel mkl? > > > of course it is and it works on all machines I have tried. If you find why > it doesn't > in your case, please report. Most likely, the version of the intel compiler > found by > configure is either old ( obscure reason. > > Paolo > --- > Paolo Giannozzi, Democritos and University of Udine, Italy > > > -- shruba gangopadhyay graduate student department of chemistry university of central florida orlando, FL-32826 'friendship doubles joys and reduces sorrows by half' (Francis Bacon). -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/c5d9a081/attachment.htm From meghdad_saeedian at yahoo.com Mon May 4 11:03:51 2009 From: meghdad_saeedian at yahoo.com (meghdad saeedian) Date: Mon, 4 May 2009 02:03:51 -0700 (PDT) Subject: [Pw_forum] bulk module (orthorhombic) Message-ID: <205834.18377.qm@web31401.mail.mud.yahoo.com> Dear all, How can?I calculate the bulk module for?an orthorhombic?lattice??with espresso? Best?regards, Saeedian -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/2b48b978/attachment.htm From paulatto at sissa.it Mon May 4 11:44:13 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 4 May 2009 11:44:13 +0200 (CEST) Subject: [Pw_forum] bulk module (orthorhombic) In-Reply-To: <205834.18377.qm@web31401.mail.mud.yahoo.com> References: <205834.18377.qm@web31401.mail.mud.yahoo.com> Message-ID: <51204.78.12.161.200.1241430253.squirrel@webmail.sissa.it> On Mon, May 4, 2009 11:03, meghdad saeedian wrote: > Dear all, > > How can?I calculate the bulk module for?an orthorhombic?lattice??with > espresso? Dear Saeedian, as far as I know, there is no ready-made program in QE that can compute the Bulk modulus from, let's say, an energy-volume curve (there is ev.x in pwtools, but it can only handle cubic and hcp systems). the easier way to get the bulk modulus is, in my opinion, to do several scf calculation around the equilibrium lattice parameters (let's say less than 1% avay) and then compute: B = -V \partial p \over \partial V by hand. The system can in principle be anisotropic, so take care to perturb it along all inequivalent axis. In order to have the pressure printed you must set tpstress=.true. in the control namelist. If you want to be more sophisticated you can use some more accurate equation of state, i.e. . regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From nazari at iasbs.ac.ir Mon May 4 15:12:30 2009 From: nazari at iasbs.ac.ir (nazari at iasbs.ac.ir) Date: Mon, 4 May 2009 14:12:30 +0100 (BST) Subject: [Pw_forum] Band Structure/graphene Message-ID: <44686.213.176.122.17.1241442750.squirrel@mail.iasbs.ac.ir> Dear All, I wish to get the band structure and band gap for graphene nanoribbon. In the following you can find the input for 2*2 graphene nanoribbon. In litrature the reported band gap for such structure is in the range of 0.5-0.6. I get such value for small number of k-point but by increasing the number of kpoint the band gap decrases and this not compatible with litratue. Is there probelm in the input. regards Fariba nazari IASBS Input for small band gap: &control calculation='nscf' prefix='graphene' pseudo_dir = '/lhome/aa/espresso-4.0.2/pseudo/', outdir='/lhome/aa/tmp/' / &system ibrav= 8, celldm(1)=9.302, celldm(2)=3.864, celldm(3)=2.844, nat=12, ntyp=2, occupations='smearing', degauss=0.02, ecutwfc =30.0, ecutrho =300.0,nbnd=28, / &electrons conv_thr = 1.0d-6 mixing_beta = 0.7 / &IONS ion_dynamics = 'bfgs' / ATOMIC_SPECIES C 12.0107 C.pbe-van_bm.UPF H 1.00794 H.pbe-van_ak.UPF ATOMIC_POSITIONS {angstrom} C 1.769889031 1.322946944 -0.012817900 C 1.770489086 4.159733324 -0.022135401 C 0.539682313 3.462493413 -0.021948354 C 0.539180075 2.020040405 -0.016808553 H 1.769947891 0.229915494 -0.001367763 H 1.774772992 5.252856855 -0.017605292 C 4.231096346 1.322960286 -0.012817558 C 4.231696827 4.159712177 -0.022136595 C 3.000909807 3.462494493 -0.021949575 C 3.000394654 2.020036555 -0.016806896 H 4.231213438 0.229870885 -0.001367555 H 4.236032200 5.252906248 -0.017606218 K_POINTS 15 0.00 0.00 0.00 1 0.0173 0.0 0.00 1 0.0273 0.0 0.0 1 0.0373 0.0 0.0 1 0.0473 0.0 0.0 1 0.0573 0.0 0.0 1 0.0673 0.0 0.0 1 0.0773 0.0 0.0 1 0.0873 0.0 0.0 1 0.0973 0.0 0.0 1 0.137 0.0 0.0 1 0.173 0.0 0.00 1 0.2754 0.0 0.0 1 0.334 0.0 0.0 1 0.50 0.00 0.00 1 Input for band gap in the range of 0.5-0.6: &control calculation='nscf' prefix='graphene' pseudo_dir = '/lhome/aa/espresso-4.0.2/pseudo/', outdir='/lhome/aa/tmp/' / &system ibrav= 8, celldm(1)=9.302, celldm(2)=3.864, celldm(3)=2.844, nat=12, ntyp=2, occupations='smearing', degauss=0.02, ecutwfc =30.0, ecutrho =300.0,nbnd=28, / &electrons conv_thr = 1.0d-6 mixing_beta = 0.7 / &IONS ion_dynamics = 'bfgs' / ATOMIC_SPECIES C 12.0107 C.pbe-van_bm.UPF H 1.00794 H.pbe-van_ak.UPF ATOMIC_POSITIONS {angstrom} C 1.769889031 1.322946944 -0.012817900 C 1.770489086 4.159733324 -0.022135401 C 0.539682313 3.462493413 -0.021948354 C 0.539180075 2.020040405 -0.016808553 H 1.769947891 0.229915494 -0.001367763 H 1.774772992 5.252856855 -0.017605292 C 4.231096346 1.322960286 -0.012817558 C 4.231696827 4.159712177 -0.022136595 C 3.000909807 3.462494493 -0.021949575 C 3.000394654 2.020036555 -0.016806896 H 4.231213438 0.229870885 -0.001367555 H 4.236032200 5.252906248 -0.017606218 K_POINTS 5 0.00 0.00 0.00 1 0.137 0.0 0.0 1 0.173 0.0 0.00 1 0.2754 0.0 0.0 1 0.50 0.00 0.00 1 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/85e7c7ae/attachment.htm From yccheng.nju at gmail.com Mon May 4 13:57:41 2009 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Mon, 4 May 2009 19:57:41 +0800 Subject: [Pw_forum] bulk module (orthorhombic) In-Reply-To: <205834.18377.qm@web31401.mail.mud.yahoo.com> References: <205834.18377.qm@web31401.mail.mud.yahoo.com> Message-ID: Dear Saeedian, If you are just interested in the bulk modulus, you can do a "vc-relax" calculation with "press='a series a value'". Then you can use the ev.x program to get the bulk modulus. But if you want to get the elastic constants, as far as I know, PWscf can not do it straightly. 2009/5/4 meghdad saeedian > Dear all, > > How can I calculate the bulk module for an *orthorhombic *lattice* * with > espresso? > > Best regards, > Saeedian > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Y. C. Cheng Department of Physics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/0b7f9024/attachment-0001.htm From Giovanni.Cantele at na.infn.it Mon May 4 15:32:35 2009 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Mon, 04 May 2009 15:32:35 +0200 Subject: [Pw_forum] Band Structure/graphene In-Reply-To: <44686.213.176.122.17.1241442750.squirrel@mail.iasbs.ac.ir> References: <44686.213.176.122.17.1241442750.squirrel@mail.iasbs.ac.ir> Message-ID: <49FEEE73.9090500@na.infn.it> nazari at iasbs.ac.ir wrote: > > Dear All, > > I wish to get the band structure and band gap for graphene nanoribbon. > In the following you can find the input for 2*2 graphene nanoribbon. > In litrature the reported band gap for such structure is in the range > of 0.5-0.6. I get such value for small number of k-point but by > increasing the number of kpoint the band gap decrases and this not > compatible with litratue. Is there probelm in the input. > > regards > Fariba nazari > IASBS > > Input for small band gap: > > &control > calculation='nscf' > prefix='graphene' > pseudo_dir = '/lhome/aa/espresso-4.0.2/pseudo/', > outdir='/lhome/aa/tmp/' > / > &system > ibrav= 8, celldm(1)=9.302, celldm(2)=3.864, celldm(3)=2.844, nat=12, > ntyp=2, > occupations='smearing', degauss=0.02, ecutwfc =30.0, ecutrho > =300.0,nbnd=28, > / > &electrons > conv_thr = 1.0d-6 > mixing_beta = 0.7 > / > &IONS > ion_dynamics = 'bfgs' > / > ATOMIC_SPECIES > C 12.0107 C.pbe-van_bm.UPF > H 1.00794 H.pbe-van_ak.UPF > ATOMIC_POSITIONS {angstrom} > C 1.769889031 1.322946944 -0.012817900 > C 1.770489086 4.159733324 -0.022135401 > C 0.539682313 3.462493413 -0.021948354 > C 0.539180075 2.020040405 -0.016808553 > H 1.769947891 0.229915494 -0.001367763 > H 1.774772992 5.252856855 -0.017605292 > C 4.231096346 1.322960286 -0.012817558 > C 4.231696827 4.159712177 -0.022136595 > C 3.000909807 3.462494493 -0.021949575 > C 3.000394654 2.020036555 -0.016806896 > H 4.231213438 0.229870885 -0.001367555 > H 4.236032200 5.252906248 -0.017606218 > K_POINTS > 15 > 0.00 0.00 0.00 1 > 0.0173 0.0 0.00 1 > 0.0273 0.0 0.0 1 > 0.0373 0.0 0.0 1 > 0.0473 0.0 0.0 1 > 0.0573 0.0 0.0 1 > 0.0673 0.0 0.0 1 > 0.0773 0.0 0.0 1 > 0.0873 0.0 0.0 1 > 0.0973 0.0 0.0 1 > 0.137 0.0 0.0 1 > 0.173 0.0 0.00 1 > 0.2754 0.0 0.0 1 > 0.334 0.0 0.0 1 > 0.50 0.00 0.00 1 > Input for band gap in the range of 0.5-0.6: > > > &control > calculation='nscf' > prefix='graphene' > pseudo_dir = '/lhome/aa/espresso-4.0.2/pseudo/', > outdir='/lhome/aa/tmp/' > / > &system > ibrav= 8, celldm(1)=9.302, celldm(2)=3.864, celldm(3)=2.844, nat=12, > ntyp=2, > occupations='smearing', degauss=0.02, ecutwfc =30.0, ecutrho > =300.0,nbnd=28, > / > &electrons > conv_thr = 1.0d-6 > mixing_beta = 0.7 > / > &IONS > ion_dynamics = 'bfgs' > / > ATOMIC_SPECIES > C 12.0107 C.pbe-van_bm.UPF > H 1.00794 H.pbe-van_ak.UPF > ATOMIC_POSITIONS {angstrom} > C 1.769889031 1.322946944 -0.012817900 > C 1.770489086 4.159733324 -0.022135401 > C 0.539682313 3.462493413 -0.021948354 > C 0.539180075 2.020040405 -0.016808553 > H 1.769947891 0.229915494 -0.001367763 > H 1.774772992 5.252856855 -0.017605292 > C 4.231096346 1.322960286 -0.012817558 > C 4.231696827 4.159712177 -0.022136595 > C 3.000909807 3.462494493 -0.021949575 > C 3.000394654 2.020036555 -0.016806896 > H 4.231213438 0.229870885 -0.001367555 > H 4.236032200 5.252906248 -0.017606218 > K_POINTS > 5 > 0.00 0.00 0.00 1 > 0.137 0.0 0.0 1 > 0.173 0.0 0.00 1 > 0.2754 0.0 0.0 1 > 0.50 0.00 0.00 1 > Hi, are you sure that the literature results you mention are just referred to this ribbon? As far as I understand, you are considering a graphene nanoribbon with zigzag edge. The literature results (e.g. Son, PRL (2006)) report that (not very aware about what happens for the width of your ribbon, but for sure this holds for a larger width): i) the spin unpolarized calculation gives a semimetal, so no gap at all (bands around the Fermi level merge from about (2/3) pi/a to pi/a, this is consistent with what you're finding!) ii) the lowest-energy configuration is spin polarized, with antiferromagnetic ordering between the two ribbon edges. Such configuration has a gap, of the order of 0.1-0.3 eV (if I well remember), but this is not the "gap" you are speaking about, being (as far as I see from your input) your calc. spin unpolarized. Therefore, it seems that your results are reasonable, the "0.5-0.6 eV gap" calc. gives a different result just because you are not considering enough k points in your path (also consider that your unit cell seems to be doubled, so the band structure is folded with respect to the ones you may found in the literature). Among the possible solutions: i) there is something "peculiar" with the zigzag-edge nanoribbon with minimum width like the one you are studying, in this case can you please also give a reference about the papers reporting the results disagreeing with yours? ii) the all-semiconducting ribbons reported in the literature are with armchair edge, are you maybe trying to reproduce results for these ribbons? In this case it seems that your unit cell reproduces the "wrong" edge. iii) did you check convergence of your scf or relax calc. with respect to parameters like: k-point grid, degauss (maybe 0.02 too large??), etc.? actually, I think that, unless one uses very unreasonable parameters, this should not change, at least qualitatively, the results. Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it From jasius_1 at yahoo.com Mon May 4 17:51:26 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Mon, 4 May 2009 08:51:26 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <627e0ffa0905040004j72f19d2fk7f8befa300ad06be@mail.gmail.com> Message-ID: <297190.3959.qm@web54404.mail.yahoo.com> I just successfully calculated DOS and PDOS of a very large system. One thing that confuses me is the presence of s, p components to PDOS for every atom, but not summed (or averaged) s and p contribution of all system. I would be happy to find in my output total DOS, PDOS s character and PDOS p character. Is there a way to get that instead of contribution per every atom? Or how do I proceed to get s and p PDOS for all system by manually processing all of my 76 atoms PDOS outputs?.. thanks Jonas Baltrusaitis University of Iowa --- On Mon, 5/4/09, Prasenjit Ghosh wrote: > From: Prasenjit Ghosh > Subject: Re: [Pw_forum] dos > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Monday, May 4, 2009, 12:04 AM > Hi Jonas, > > For DOS calculations one needs to do an integration over > the kpts in the > Brillouin zone. So you need a very fine k-point mesh than > necessary for a > scf calculation. Moreover you need a large number of bands. > Doing an scf > calculation with such a fine k-point mesh and so many bands > is very > expensive. So you should first do an scf calculation with > the min. no. of > k-points and bands and then do a nscf calculation with a > finer k-point and > larger no. of bands. In the nscf step, the code does an > interpolation for > the extra bands and k-points and not a self-consistent > calculation. This > saves computational time. > > Hope this helps. > > Prasenjit. > > 2009/5/3 Jonas Baltrusaitis > > > > > Prasenjit, > > > > could you explain why both scf followed by nscf > calculations are needed for > > DOS calculations, as per example 8? If only scf > density is needed for > > further projection, why not to calculate only one? > > > > Jonas > > > > > > --- On Sun, 5/3/09, Prasenjit Ghosh > wrote: > > > > > From: Prasenjit Ghosh > > > > Subject: Re: [Pw_forum] dos > > > To: "PWSCF Forum" > > > > Date: Sunday, May 3, 2009, 2:00 PM > > > dear all user > > > > > > > > > > Dear Yaldaa, > > > > > > > > > > How can I project dos to molecular orbital? > > > > > > > > > > There is a post-processing tool called projwfc.x > which you > > > can use to > > > project dos to molecular orbital. > > > > > > what is the importance of nbnd in calculations > for dos? > > > > > > > e > > > "nbnd" tells the code the no. of > Kohn-Sham > > > states it has to > > > compute.....While computing the dos, one wants to > plot the > > > dos for both the > > > occupied & unoccupied manifold, the no. of > energy > > > levels is read in by the > > > dos.x from the output of your scf/nscf run. So > while doing > > > an scf/ nscf > > > calculations you can control the total no. of > states with > > > the variable > > > "nbnd". > > > Depending on the type of occupation scheme you > choose the > > > value of nbnd is > > > nelec/2 (fixed occupation) or nelec/2+some extra > bands > > > (when you use > > > smearing). If you want to plot the dos, usually > you should > > > add some extra > > > levels using nbnd & do the scf or nscf > calculations to > > > get the unoccupied > > > manifold. > > > > > > and for plottingdos or pdos, if the fermi energy > is assumed > > > to be zero > > > > automatically by program? > > > > > > > > > > No, you have to do the alignment yourself > > > > > > With regards, > > > > > > Prasenjit. > > > > > > > > > > > > -- > > > PRASENJIT GHOSH, > > > POST-DOC, > > > ROOM NO: 265, MAIN BUILDING, > > > CM SECTION, ICTP, > > > STRADA COSTERIA 11, > > > TRIESTE, 34104, > > > ITALY > > > PHONE: +39 040 2240 369 (O) > > > +39 3807528672 (M) > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > -- > PRASENJIT GHOSH, > POST-DOC, > ROOM NO: 265, MAIN BUILDING, > CM SECTION, ICTP, > STRADA COSTERIA 11, > TRIESTE, 34104, > ITALY > PHONE: +39 040 2240 369 (O) > +39 3807528672 (M) From giannozz at democritos.it Mon May 4 17:54:49 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 04 May 2009 17:54:49 +0200 Subject: [Pw_forum] dos In-Reply-To: <297190.3959.qm@web54404.mail.yahoo.com> References: <297190.3959.qm@web54404.mail.yahoo.com> Message-ID: <49FF0FC9.5050004@democritos.it> Jonas Baltrusaitis wrote: > I would be happy to find in my output total DOS, PDOS s character and PDOS > p character. Is there a way to get that instead of contribution per every atom? please have a look at PP/sumpdos.f90 . It may not do exactly what you like, but it shouldn't be difficult to modify it Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From lex at phys.ufl.edu Mon May 4 17:55:59 2009 From: lex at phys.ufl.edu (Lex Kemper) Date: Mon, 04 May 2009 11:55:59 -0400 Subject: [Pw_forum] dos In-Reply-To: <297190.3959.qm@web54404.mail.yahoo.com> References: <297190.3959.qm@web54404.mail.yahoo.com> Message-ID: <49FF100F.5030806@phys.ufl.edu> Hi Jonas, First, I don't know if a quantity such as the 'p' character of a system is a useful thing to know, unless all the atoms are the same. If you want to do so anyway, try familiarizing yourself with 'awk'. It's a very useful utility for parsing text files. Lex Kemper University of Florida Jonas Baltrusaitis wrote: > I just successfully calculated DOS and PDOS of a very large system. One thing that confuses me is the presence of s, p components to PDOS for every atom, but not summed (or averaged) s and p contribution of all system. I would be happy to find in my output total DOS, PDOS s character and PDOS p character. Is there a way to get that instead of contribution per every atom? Or how do I proceed to get s and p PDOS for all system by manually processing all of my 76 atoms PDOS outputs?.. > > thanks > > Jonas Baltrusaitis > University of Iowa > > > --- On Mon, 5/4/09, Prasenjit Ghosh wrote: > >> From: Prasenjit Ghosh >> Subject: Re: [Pw_forum] dos >> To: jasius_1 at yahoo.com, "PWSCF Forum" >> Date: Monday, May 4, 2009, 12:04 AM >> Hi Jonas, >> >> For DOS calculations one needs to do an integration over >> the kpts in the >> Brillouin zone. So you need a very fine k-point mesh than >> necessary for a >> scf calculation. Moreover you need a large number of bands. >> Doing an scf >> calculation with such a fine k-point mesh and so many bands >> is very >> expensive. So you should first do an scf calculation with >> the min. no. of >> k-points and bands and then do a nscf calculation with a >> finer k-point and >> larger no. of bands. In the nscf step, the code does an >> interpolation for >> the extra bands and k-points and not a self-consistent >> calculation. This >> saves computational time. >> >> Hope this helps. >> >> Prasenjit. >> >> 2009/5/3 Jonas Baltrusaitis >> >>> Prasenjit, >>> >>> could you explain why both scf followed by nscf >> calculations are needed for >>> DOS calculations, as per example 8? If only scf >> density is needed for >>> further projection, why not to calculate only one? >>> >>> Jonas >>> >>> >>> --- On Sun, 5/3/09, Prasenjit Ghosh >> wrote: >>>> From: Prasenjit Ghosh >> >>>> Subject: Re: [Pw_forum] dos >>>> To: "PWSCF Forum" >> >>>> Date: Sunday, May 3, 2009, 2:00 PM >>>> dear all user >>>> Dear Yaldaa, >>>> >>>> >>>>> How can I project dos to molecular orbital? >>>>> >>>> There is a post-processing tool called projwfc.x >> which you >>>> can use to >>>> project dos to molecular orbital. >>>> >>>> what is the importance of nbnd in calculations >> for dos? >>>> e >>>> "nbnd" tells the code the no. of >> Kohn-Sham >>>> states it has to >>>> compute.....While computing the dos, one wants to >> plot the >>>> dos for both the >>>> occupied & unoccupied manifold, the no. of >> energy >>>> levels is read in by the >>>> dos.x from the output of your scf/nscf run. So >> while doing >>>> an scf/ nscf >>>> calculations you can control the total no. of >> states with >>>> the variable >>>> "nbnd". >>>> Depending on the type of occupation scheme you >> choose the >>>> value of nbnd is >>>> nelec/2 (fixed occupation) or nelec/2+some extra >> bands >>>> (when you use >>>> smearing). If you want to plot the dos, usually >> you should >>>> add some extra >>>> levels using nbnd & do the scf or nscf >> calculations to >>>> get the unoccupied >>>> manifold. >>>> >>>> and for plottingdos or pdos, if the fermi energy >> is assumed >>>> to be zero >>>>> automatically by program? >>>>> >>>> No, you have to do the alignment yourself >>>> >>>> With regards, >>>> >>>> Prasenjit. >>>> >>>> >>>> >>>> -- >>>> PRASENJIT GHOSH, >>>> POST-DOC, >>>> ROOM NO: 265, MAIN BUILDING, >>>> CM SECTION, ICTP, >>>> STRADA COSTERIA 11, >>>> TRIESTE, 34104, >>>> ITALY >>>> PHONE: +39 040 2240 369 (O) >>>> +39 3807528672 (M) >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> >> http://www.democritos.it/mailman/listinfo/pw_forum >>> >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> >> >> -- >> PRASENJIT GHOSH, >> POST-DOC, >> ROOM NO: 265, MAIN BUILDING, >> CM SECTION, ICTP, >> STRADA COSTERIA 11, >> TRIESTE, 34104, >> ITALY >> PHONE: +39 040 2240 369 (O) >> +39 3807528672 (M) > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From baroni at sissa.it Mon May 4 17:55:28 2009 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 4 May 2009 17:55:28 +0200 Subject: [Pw_forum] dos In-Reply-To: <297190.3959.qm@web54404.mail.yahoo.com> References: <297190.3959.qm@web54404.mail.yahoo.com> Message-ID: <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> Jonas: I am afraid that what you want to obtain does not make much sense. The angular character of any molecular (or Bloch) state depends on the origin of your reference system. What is "s-like" with respect to one point has an infinite number of angular components with respect to a different point. Are you sure you want to add "what is s-like in Iowa" with "what is s-like in Italy" ??? ;-) Stefano On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > > I just successfully calculated DOS and PDOS of a very large system. > One thing that confuses me is the presence of s, p components to > PDOS for every atom, but not summed (or averaged) s and p > contribution of all system. I would be happy to find in my output > total DOS, PDOS s character and PDOS p character. Is there a way to > get that instead of contribution per every atom? Or how do I proceed > to get s and p PDOS for all system by manually processing all of my > 76 atoms PDOS outputs?.. > > thanks > > Jonas Baltrusaitis > University of Iowa > > > --- On Mon, 5/4/09, Prasenjit Ghosh wrote: > >> From: Prasenjit Ghosh >> Subject: Re: [Pw_forum] dos >> To: jasius_1 at yahoo.com, "PWSCF Forum" >> Date: Monday, May 4, 2009, 12:04 AM >> Hi Jonas, >> >> For DOS calculations one needs to do an integration over >> the kpts in the >> Brillouin zone. So you need a very fine k-point mesh than >> necessary for a >> scf calculation. Moreover you need a large number of bands. >> Doing an scf >> calculation with such a fine k-point mesh and so many bands >> is very >> expensive. So you should first do an scf calculation with >> the min. no. of >> k-points and bands and then do a nscf calculation with a >> finer k-point and >> larger no. of bands. In the nscf step, the code does an >> interpolation for >> the extra bands and k-points and not a self-consistent >> calculation. This >> saves computational time. >> >> Hope this helps. >> >> Prasenjit. >> >> 2009/5/3 Jonas Baltrusaitis >> >>> >>> Prasenjit, >>> >>> could you explain why both scf followed by nscf >> calculations are needed for >>> DOS calculations, as per example 8? If only scf >> density is needed for >>> further projection, why not to calculate only one? >>> >>> Jonas >>> >>> >>> --- On Sun, 5/3/09, Prasenjit Ghosh >> wrote: >>> >>>> From: Prasenjit Ghosh >> >>>> Subject: Re: [Pw_forum] dos >>>> To: "PWSCF Forum" >> >>>> Date: Sunday, May 3, 2009, 2:00 PM >>>> dear all user >>>>> >>>> >>>> Dear Yaldaa, >>>> >>>> >>>>> How can I project dos to molecular orbital? >>>>> >>>> >>>> There is a post-processing tool called projwfc.x >> which you >>>> can use to >>>> project dos to molecular orbital. >>>> >>>> what is the importance of nbnd in calculations >> for dos? >>>>> >>>> e >>>> "nbnd" tells the code the no. of >> Kohn-Sham >>>> states it has to >>>> compute.....While computing the dos, one wants to >> plot the >>>> dos for both the >>>> occupied & unoccupied manifold, the no. of >> energy >>>> levels is read in by the >>>> dos.x from the output of your scf/nscf run. So >> while doing >>>> an scf/ nscf >>>> calculations you can control the total no. of >> states with >>>> the variable >>>> "nbnd". >>>> Depending on the type of occupation scheme you >> choose the >>>> value of nbnd is >>>> nelec/2 (fixed occupation) or nelec/2+some extra >> bands >>>> (when you use >>>> smearing). If you want to plot the dos, usually >> you should >>>> add some extra >>>> levels using nbnd & do the scf or nscf >> calculations to >>>> get the unoccupied >>>> manifold. >>>> >>>> and for plottingdos or pdos, if the fermi energy >> is assumed >>>> to be zero >>>>> automatically by program? >>>>> >>>> >>>> No, you have to do the alignment yourself >>>> >>>> With regards, >>>> >>>> Prasenjit. >>>> >>>> >>>> >>>> -- >>>> PRASENJIT GHOSH, >>>> POST-DOC, >>>> ROOM NO: 265, MAIN BUILDING, >>>> CM SECTION, ICTP, >>>> STRADA COSTERIA 11, >>>> TRIESTE, 34104, >>>> ITALY >>>> PHONE: +39 040 2240 369 (O) >>>> +39 3807528672 (M) >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> >> http://www.democritos.it/mailman/listinfo/pw_forum >>> >>> >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> >> >> >> -- >> PRASENJIT GHOSH, >> POST-DOC, >> ROOM NO: 265, MAIN BUILDING, >> CM SECTION, ICTP, >> STRADA COSTERIA 11, >> TRIESTE, 34104, >> ITALY >> PHONE: +39 040 2240 369 (O) >> +39 3807528672 (M) > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/4eadd31d/attachment-0001.htm From baroni at sissa.it Mon May 4 17:57:07 2009 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 4 May 2009 17:57:07 +0200 Subject: [Pw_forum] dos In-Reply-To: <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> Message-ID: <93543A94-94EC-4DF3-AE45-5655971173A1@sissa.it> ... and pardon us if it took eight minutes to get three decent answers ... Enjoy! Stefano On May 4, 2009, at 5:55 PM, Stefano Baroni wrote: > Jonas: I am afraid that what you want to obtain does not make much > sense. The angular character of any molecular (or Bloch) state > depends on the origin of your reference system. What is "s-like" > with respect to one point has an infinite number of angular > components with respect to a different point. Are you sure you want > to add "what is s-like in Iowa" with "what is s-like in Italy" ??? ;-) > Stefano > > On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > >> >> I just successfully calculated DOS and PDOS of a very large system. >> One thing that confuses me is the presence of s, p components to >> PDOS for every atom, but not summed (or averaged) s and p >> contribution of all system. I would be happy to find in my output >> total DOS, PDOS s character and PDOS p character. Is there a way to >> get that instead of contribution per every atom? Or how do I >> proceed to get s and p PDOS for all system by manually processing >> all of my 76 atoms PDOS outputs?.. >> >> thanks >> >> Jonas Baltrusaitis >> University of Iowa >> >> >> --- On Mon, 5/4/09, Prasenjit Ghosh wrote: >> >>> From: Prasenjit Ghosh >>> Subject: Re: [Pw_forum] dos >>> To: jasius_1 at yahoo.com, "PWSCF Forum" >>> Date: Monday, May 4, 2009, 12:04 AM >>> Hi Jonas, >>> >>> For DOS calculations one needs to do an integration over >>> the kpts in the >>> Brillouin zone. So you need a very fine k-point mesh than >>> necessary for a >>> scf calculation. Moreover you need a large number of bands. >>> Doing an scf >>> calculation with such a fine k-point mesh and so many bands >>> is very >>> expensive. So you should first do an scf calculation with >>> the min. no. of >>> k-points and bands and then do a nscf calculation with a >>> finer k-point and >>> larger no. of bands. In the nscf step, the code does an >>> interpolation for >>> the extra bands and k-points and not a self-consistent >>> calculation. This >>> saves computational time. >>> >>> Hope this helps. >>> >>> Prasenjit. >>> >>> 2009/5/3 Jonas Baltrusaitis >>> >>>> >>>> Prasenjit, >>>> >>>> could you explain why both scf followed by nscf >>> calculations are needed for >>>> DOS calculations, as per example 8? If only scf >>> density is needed for >>>> further projection, why not to calculate only one? >>>> >>>> Jonas >>>> >>>> >>>> --- On Sun, 5/3/09, Prasenjit Ghosh >>> wrote: >>>> >>>>> From: Prasenjit Ghosh >>> >>>>> Subject: Re: [Pw_forum] dos >>>>> To: "PWSCF Forum" >>> >>>>> Date: Sunday, May 3, 2009, 2:00 PM >>>>> dear all user >>>>>> >>>>> >>>>> Dear Yaldaa, >>>>> >>>>> >>>>>> How can I project dos to molecular orbital? >>>>>> >>>>> >>>>> There is a post-processing tool called projwfc.x >>> which you >>>>> can use to >>>>> project dos to molecular orbital. >>>>> >>>>> what is the importance of nbnd in calculations >>> for dos? >>>>>> >>>>> e >>>>> "nbnd" tells the code the no. of >>> Kohn-Sham >>>>> states it has to >>>>> compute.....While computing the dos, one wants to >>> plot the >>>>> dos for both the >>>>> occupied & unoccupied manifold, the no. of >>> energy >>>>> levels is read in by the >>>>> dos.x from the output of your scf/nscf run. So >>> while doing >>>>> an scf/ nscf >>>>> calculations you can control the total no. of >>> states with >>>>> the variable >>>>> "nbnd". >>>>> Depending on the type of occupation scheme you >>> choose the >>>>> value of nbnd is >>>>> nelec/2 (fixed occupation) or nelec/2+some extra >>> bands >>>>> (when you use >>>>> smearing). If you want to plot the dos, usually >>> you should >>>>> add some extra >>>>> levels using nbnd & do the scf or nscf >>> calculations to >>>>> get the unoccupied >>>>> manifold. >>>>> >>>>> and for plottingdos or pdos, if the fermi energy >>> is assumed >>>>> to be zero >>>>>> automatically by program? >>>>>> >>>>> >>>>> No, you have to do the alignment yourself >>>>> >>>>> With regards, >>>>> >>>>> Prasenjit. >>>>> >>>>> >>>>> >>>>> -- >>>>> PRASENJIT GHOSH, >>>>> POST-DOC, >>>>> ROOM NO: 265, MAIN BUILDING, >>>>> CM SECTION, ICTP, >>>>> STRADA COSTERIA 11, >>>>> TRIESTE, 34104, >>>>> ITALY >>>>> PHONE: +39 040 2240 369 (O) >>>>> +39 3807528672 (M) >>>>> _______________________________________________ >>>>> Pw_forum mailing list >>>>> Pw_forum at pwscf.org >>>>> >>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>>> >>>> >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>> >>> >>> >>> -- >>> PRASENJIT GHOSH, >>> POST-DOC, >>> ROOM NO: 265, MAIN BUILDING, >>> CM SECTION, ICTP, >>> STRADA COSTERIA 11, >>> TRIESTE, 34104, >>> ITALY >>> PHONE: +39 040 2240 369 (O) >>> +39 3807528672 (M) >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - > Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une > morale de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090504/6cda0cf3/attachment-0001.htm From jasius_1 at yahoo.com Mon May 4 18:28:52 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Mon, 4 May 2009 09:28:52 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> Message-ID: <128832.24342.qm@web54410.mail.yahoo.com> that's a very good point. so far in the literature I've seen plots only of TDOS and PDOS of s, p and d components. I have never seen plotting components of a single atom. I just googled the first available reference PWSCF DOS calculations: Koichiro Umemoto1, Renata M. Wentzcovitch, Lan Wang1, and Chris Leighton, Electronic structure of Co1?xFexS2, phys. stat. sol. (b) 243, No. 9, 2117?2121 (2006). In page 2118 there is a simple DOS and PDOS for CoS2, plotted as TDOS, PDOS Co3d, PDOS S3s and S3p. That is what I would like to get. I am sure they got s, p and d components for every single atom, too. The question is how it can be "averaged" And yes, I will look at two other suggestions I appreciate your help Jonas --- On Mon, 5/4/09, Stefano Baroni wrote: > From: Stefano Baroni > Subject: Re: [Pw_forum] dos > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Monday, May 4, 2009, 8:55 AM > Jonas: I am afraid that what you want to obtain does not > make much sense. The angular character of any molecular (or > Bloch) state depends on the origin of your reference system. > What is "s-like" with respect to one point has an > infinite number of angular components with respect to a > different point. Are you sure you want to add "what is > s-like in Iowa" with "what is s-like in > Italy" ??? ;-) > Stefano > > On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > > > > > I just successfully calculated DOS and PDOS of a very > large system. One thing that confuses me is the presence of > s, p components to PDOS for every atom, but not summed (or > averaged) s and p contribution of all system. I would be > happy to find in my output total DOS, PDOS s character and > PDOS p character. Is there a way to get that instead of > contribution per every atom? Or how do I proceed to get s > and p PDOS for all system by manually processing all of my > 76 atoms PDOS outputs?.. > > > > thanks > > > > Jonas Baltrusaitis > > University of Iowa > > > > > > --- On Mon, 5/4/09, Prasenjit Ghosh > wrote: > > > >> From: Prasenjit Ghosh > > >> Subject: Re: [Pw_forum] dos > >> To: jasius_1 at yahoo.com, "PWSCF Forum" > > >> Date: Monday, May 4, 2009, 12:04 AM > >> Hi Jonas, > >> > >> For DOS calculations one needs to do an > integration over > >> the kpts in the > >> Brillouin zone. So you need a very fine k-point > mesh than > >> necessary for a > >> scf calculation. Moreover you need a large number > of bands. > >> Doing an scf > >> calculation with such a fine k-point mesh and so > many bands > >> is very > >> expensive. So you should first do an scf > calculation with > >> the min. no. of > >> k-points and bands and then do a nscf calculation > with a > >> finer k-point and > >> larger no. of bands. In the nscf step, the code > does an > >> interpolation for > >> the extra bands and k-points and not a > self-consistent > >> calculation. This > >> saves computational time. > >> > >> Hope this helps. > >> > >> Prasenjit. > >> > >> 2009/5/3 Jonas Baltrusaitis > > >> > >>> > >>> Prasenjit, > >>> > >>> could you explain why both scf followed by > nscf > >> calculations are needed for > >>> DOS calculations, as per example 8? If only > scf > >> density is needed for > >>> further projection, why not to calculate only > one? > >>> > >>> Jonas > >>> > >>> > >>> --- On Sun, 5/3/09, Prasenjit Ghosh > >> wrote: > >>> > >>>> From: Prasenjit Ghosh > >> > >>>> Subject: Re: [Pw_forum] dos > >>>> To: "PWSCF Forum" > >> > >>>> Date: Sunday, May 3, 2009, 2:00 PM > >>>> dear all user > >>>>> > >>>> > >>>> Dear Yaldaa, > >>>> > >>>> > >>>>> How can I project dos to molecular > orbital? > >>>>> > >>>> > >>>> There is a post-processing tool called > projwfc.x > >> which you > >>>> can use to > >>>> project dos to molecular orbital. > >>>> > >>>> what is the importance of nbnd in > calculations > >> for dos? > >>>>> > >>>> e > >>>> "nbnd" tells the code the no. > of > >> Kohn-Sham > >>>> states it has to > >>>> compute.....While computing the dos, one > wants to > >> plot the > >>>> dos for both the > >>>> occupied & unoccupied manifold, the > no. of > >> energy > >>>> levels is read in by the > >>>> dos.x from the output of your scf/nscf > run. So > >> while doing > >>>> an scf/ nscf > >>>> calculations you can control the total no. > of > >> states with > >>>> the variable > >>>> "nbnd". > >>>> Depending on the type of occupation scheme > you > >> choose the > >>>> value of nbnd is > >>>> nelec/2 (fixed occupation) or > nelec/2+some extra > >> bands > >>>> (when you use > >>>> smearing). If you want to plot the dos, > usually > >> you should > >>>> add some extra > >>>> levels using nbnd & do the scf or nscf > >> calculations to > >>>> get the unoccupied > >>>> manifold. > >>>> > >>>> and for plottingdos or pdos, if the fermi > energy > >> is assumed > >>>> to be zero > >>>>> automatically by program? > >>>>> > >>>> > >>>> No, you have to do the alignment yourself > >>>> > >>>> With regards, > >>>> > >>>> Prasenjit. > >>>> > >>>> > >>>> > >>>> -- > >>>> PRASENJIT GHOSH, > >>>> POST-DOC, > >>>> ROOM NO: 265, MAIN BUILDING, > >>>> CM SECTION, ICTP, > >>>> STRADA COSTERIA 11, > >>>> TRIESTE, 34104, > >>>> ITALY > >>>> PHONE: +39 040 2240 369 (O) > >>>> +39 3807528672 (M) > >>>> > _______________________________________________ > >>>> Pw_forum mailing list > >>>> Pw_forum at pwscf.org > >>>> > >> http://www.democritos.it/mailman/listinfo/pw_forum > >>> > >>> > >>> > >>> > _______________________________________________ > >>> Pw_forum mailing list > >>> Pw_forum at pwscf.org > >>> > http://www.democritos.it/mailman/listinfo/pw_forum > >>> > >> > >> > >> > >> --PRASENJIT GHOSH, > >> POST-DOC, > >> ROOM NO: 265, MAIN BUILDING, > >> CM SECTION, ICTP, > >> STRADA COSTERIA 11, > >> TRIESTE, 34104, > >> ITALY > >> PHONE: +39 040 2240 369 (O) > >> +39 3807528672 (M) > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Stefano Baroni - SISSA & DEMOCRITOS National > Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > La morale est une logique de l'action comme la logique > est une morale de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or > PowerPoint attachments > Why? See: > http://www.gnu.org/philosophy/no-word-attachments.html From jasius_1 at yahoo.com Mon May 4 18:31:50 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Mon, 4 May 2009 09:31:50 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <49FF100F.5030806@phys.ufl.edu> Message-ID: <858222.68922.qm@web54409.mail.yahoo.com> I have N,C,H and O. I would like to have them plotted as O2s, O2p, C2s, C2p etc. That would make sense, I guess. I don't want to sum N and C2s contribution, for example Jonas --- On Mon, 5/4/09, Lex Kemper wrote: > From: Lex Kemper > Subject: Re: [Pw_forum] dos > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Monday, May 4, 2009, 8:55 AM > Hi Jonas, > > First, I don't know if a quantity such as the > 'p' character of a system > is a useful thing to know, unless all the atoms are the > same. If you > want to do so anyway, try familiarizing yourself with > 'awk'. It's a very > useful utility for parsing text files. > > Lex Kemper > University of Florida > > Jonas Baltrusaitis wrote: > > I just successfully calculated DOS and PDOS of a very > large system. One thing that confuses me is the presence of > s, p components to PDOS for every atom, but not summed (or > averaged) s and p contribution of all system. I would be > happy to find in my output total DOS, PDOS s character and > PDOS p character. Is there a way to get that instead of > contribution per every atom? Or how do I proceed to get s > and p PDOS for all system by manually processing all of my > 76 atoms PDOS outputs?.. > > > > thanks > > > > Jonas Baltrusaitis > > University of Iowa > > > > > > --- On Mon, 5/4/09, Prasenjit Ghosh > wrote: > > > >> From: Prasenjit Ghosh > > >> Subject: Re: [Pw_forum] dos > >> To: jasius_1 at yahoo.com, "PWSCF Forum" > > >> Date: Monday, May 4, 2009, 12:04 AM > >> Hi Jonas, > >> > >> For DOS calculations one needs to do an > integration over > >> the kpts in the > >> Brillouin zone. So you need a very fine k-point > mesh than > >> necessary for a > >> scf calculation. Moreover you need a large number > of bands. > >> Doing an scf > >> calculation with such a fine k-point mesh and so > many bands > >> is very > >> expensive. So you should first do an scf > calculation with > >> the min. no. of > >> k-points and bands and then do a nscf calculation > with a > >> finer k-point and > >> larger no. of bands. In the nscf step, the code > does an > >> interpolation for > >> the extra bands and k-points and not a > self-consistent > >> calculation. This > >> saves computational time. > >> > >> Hope this helps. > >> > >> Prasenjit. > >> > >> 2009/5/3 Jonas Baltrusaitis > > >> > >>> Prasenjit, > >>> > >>> could you explain why both scf followed by > nscf > >> calculations are needed for > >>> DOS calculations, as per example 8? If only > scf > >> density is needed for > >>> further projection, why not to calculate only > one? > >>> > >>> Jonas > >>> > >>> > >>> --- On Sun, 5/3/09, Prasenjit Ghosh > >> wrote: > >>>> From: Prasenjit Ghosh > >> > >>>> Subject: Re: [Pw_forum] dos > >>>> To: "PWSCF Forum" > >> > >>>> Date: Sunday, May 3, 2009, 2:00 PM > >>>> dear all user > >>>> Dear Yaldaa, > >>>> > >>>> > >>>>> How can I project dos to molecular > orbital? > >>>>> > >>>> There is a post-processing tool called > projwfc.x > >> which you > >>>> can use to > >>>> project dos to molecular orbital. > >>>> > >>>> what is the importance of nbnd in > calculations > >> for dos? > >>>> e > >>>> "nbnd" tells the code the no. > of > >> Kohn-Sham > >>>> states it has to > >>>> compute.....While computing the dos, one > wants to > >> plot the > >>>> dos for both the > >>>> occupied & unoccupied manifold, the > no. of > >> energy > >>>> levels is read in by the > >>>> dos.x from the output of your scf/nscf > run. So > >> while doing > >>>> an scf/ nscf > >>>> calculations you can control the total no. > of > >> states with > >>>> the variable > >>>> "nbnd". > >>>> Depending on the type of occupation scheme > you > >> choose the > >>>> value of nbnd is > >>>> nelec/2 (fixed occupation) or > nelec/2+some extra > >> bands > >>>> (when you use > >>>> smearing). If you want to plot the dos, > usually > >> you should > >>>> add some extra > >>>> levels using nbnd & do the scf or nscf > >> calculations to > >>>> get the unoccupied > >>>> manifold. > >>>> > >>>> and for plottingdos or pdos, if the fermi > energy > >> is assumed > >>>> to be zero > >>>>> automatically by program? > >>>>> > >>>> No, you have to do the alignment yourself > >>>> > >>>> With regards, > >>>> > >>>> Prasenjit. > >>>> > >>>> > >>>> > >>>> -- > >>>> PRASENJIT GHOSH, > >>>> POST-DOC, > >>>> ROOM NO: 265, MAIN BUILDING, > >>>> CM SECTION, ICTP, > >>>> STRADA COSTERIA 11, > >>>> TRIESTE, 34104, > >>>> ITALY > >>>> PHONE: +39 040 2240 369 (O) > >>>> +39 3807528672 (M) > >>>> > _______________________________________________ > >>>> Pw_forum mailing list > >>>> Pw_forum at pwscf.org > >>>> > >> http://www.democritos.it/mailman/listinfo/pw_forum > >>> > >>> > >>> > _______________________________________________ > >>> Pw_forum mailing list > >>> Pw_forum at pwscf.org > >>> > http://www.democritos.it/mailman/listinfo/pw_forum > >>> > >> > >> > >> -- > >> PRASENJIT GHOSH, > >> POST-DOC, > >> ROOM NO: 265, MAIN BUILDING, > >> CM SECTION, ICTP, > >> STRADA COSTERIA 11, > >> TRIESTE, 34104, > >> ITALY > >> PHONE: +39 040 2240 369 (O) > >> +39 3807528672 (M) > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum From jasius_1 at yahoo.com Mon May 4 19:23:34 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Mon, 4 May 2009 10:23:34 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <93543A94-94EC-4DF3-AE45-5655971173A1@sissa.it> Message-ID: <908110.60693.qm@web54407.mail.yahoo.com> yep, the reply was swift. that's why I come to this forum --- On Mon, 5/4/09, Stefano Baroni wrote: > From: Stefano Baroni > Subject: Re: [Pw_forum] dos > To: "PWSCF Forum" > Date: Monday, May 4, 2009, 8:57 AM > ... and pardon us if it took eight minutes to get three > decent answers ... > Enjoy! Stefano > > On May 4, 2009, at 5:55 PM, Stefano Baroni wrote: > > > Jonas: I am afraid that what you want to obtain does > not make much sense. The angular character of any molecular > (or Bloch) state depends on the origin of your reference > system. What is "s-like" with respect to one point > has an infinite number of angular components with respect to > a different point. Are you sure you want to add "what > is s-like in Iowa" with "what is s-like in > Italy" ??? ;-) > > Stefano > > > > On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > > > >> > >> I just successfully calculated DOS and PDOS of a > very large system. One thing that confuses me is the > presence of s, p components to PDOS for every atom, but not > summed (or averaged) s and p contribution of all system. I > would be happy to find in my output total DOS, PDOS s > character and PDOS p character. Is there a way to get that > instead of contribution per every atom? Or how do I proceed > to get s and p PDOS for all system by manually processing > all of my 76 atoms PDOS outputs?.. > >> > >> thanks > >> > >> Jonas Baltrusaitis > >> University of Iowa > >> > >> > >> --- On Mon, 5/4/09, Prasenjit Ghosh > wrote: > >> > >>> From: Prasenjit Ghosh > > >>> Subject: Re: [Pw_forum] dos > >>> To: jasius_1 at yahoo.com, "PWSCF > Forum" > >>> Date: Monday, May 4, 2009, 12:04 AM > >>> Hi Jonas, > >>> > >>> For DOS calculations one needs to do an > integration over > >>> the kpts in the > >>> Brillouin zone. So you need a very fine > k-point mesh than > >>> necessary for a > >>> scf calculation. Moreover you need a large > number of bands. > >>> Doing an scf > >>> calculation with such a fine k-point mesh and > so many bands > >>> is very > >>> expensive. So you should first do an scf > calculation with > >>> the min. no. of > >>> k-points and bands and then do a nscf > calculation with a > >>> finer k-point and > >>> larger no. of bands. In the nscf step, the > code does an > >>> interpolation for > >>> the extra bands and k-points and not a > self-consistent > >>> calculation. This > >>> saves computational time. > >>> > >>> Hope this helps. > >>> > >>> Prasenjit. > >>> > >>> 2009/5/3 Jonas Baltrusaitis > > >>> > >>>> > >>>> Prasenjit, > >>>> > >>>> could you explain why both scf followed by > nscf > >>> calculations are needed for > >>>> DOS calculations, as per example 8? If > only scf > >>> density is needed for > >>>> further projection, why not to calculate > only one? > >>>> > >>>> Jonas > >>>> > >>>> > >>>> --- On Sun, 5/3/09, Prasenjit Ghosh > >>> wrote: > >>>> > >>>>> From: Prasenjit Ghosh > >>> > >>>>> Subject: Re: [Pw_forum] dos > >>>>> To: "PWSCF Forum" > >>> > >>>>> Date: Sunday, May 3, 2009, 2:00 PM > >>>>> dear all user > >>>>>> > >>>>> > >>>>> Dear Yaldaa, > >>>>> > >>>>> > >>>>>> How can I project dos to molecular > orbital? > >>>>>> > >>>>> > >>>>> There is a post-processing tool called > projwfc.x > >>> which you > >>>>> can use to > >>>>> project dos to molecular orbital. > >>>>> > >>>>> what is the importance of nbnd in > calculations > >>> for dos? > >>>>>> > >>>>> e > >>>>> "nbnd" tells the code the > no. of > >>> Kohn-Sham > >>>>> states it has to > >>>>> compute.....While computing the dos, > one wants to > >>> plot the > >>>>> dos for both the > >>>>> occupied & unoccupied manifold, > the no. of > >>> energy > >>>>> levels is read in by the > >>>>> dos.x from the output of your scf/nscf > run. So > >>> while doing > >>>>> an scf/ nscf > >>>>> calculations you can control the total > no. of > >>> states with > >>>>> the variable > >>>>> "nbnd". > >>>>> Depending on the type of occupation > scheme you > >>> choose the > >>>>> value of nbnd is > >>>>> nelec/2 (fixed occupation) or > nelec/2+some extra > >>> bands > >>>>> (when you use > >>>>> smearing). If you want to plot the > dos, usually > >>> you should > >>>>> add some extra > >>>>> levels using nbnd & do the scf or > nscf > >>> calculations to > >>>>> get the unoccupied > >>>>> manifold. > >>>>> > >>>>> and for plottingdos or pdos, if the > fermi energy > >>> is assumed > >>>>> to be zero > >>>>>> automatically by program? > >>>>>> > >>>>> > >>>>> No, you have to do the alignment > yourself > >>>>> > >>>>> With regards, > >>>>> > >>>>> Prasenjit. > >>>>> > >>>>> > >>>>> > >>>>> -- > >>>>> PRASENJIT GHOSH, > >>>>> POST-DOC, > >>>>> ROOM NO: 265, MAIN BUILDING, > >>>>> CM SECTION, ICTP, > >>>>> STRADA COSTERIA 11, > >>>>> TRIESTE, 34104, > >>>>> ITALY > >>>>> PHONE: +39 040 2240 369 (O) > >>>>> +39 3807528672 (M) > >>>>> > _______________________________________________ > >>>>> Pw_forum mailing list > >>>>> Pw_forum at pwscf.org > >>>>> > >>> > http://www.democritos.it/mailman/listinfo/pw_forum > >>>> > >>>> > >>>> > >>>> > _______________________________________________ > >>>> Pw_forum mailing list > >>>> Pw_forum at pwscf.org > >>>> > http://www.democritos.it/mailman/listinfo/pw_forum > >>>> > >>> > >>> > >>> > >>> --PRASENJIT GHOSH, > >>> POST-DOC, > >>> ROOM NO: 265, MAIN BUILDING, > >>> CM SECTION, ICTP, > >>> STRADA COSTERIA 11, > >>> TRIESTE, 34104, > >>> ITALY > >>> PHONE: +39 040 2240 369 (O) > >>> +39 3807528672 (M) > >> > >> > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > > Stefano Baroni - SISSA & DEMOCRITOS National > Simulation Center - Trieste > > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni > (skype) > > > > La morale est une logique de l'action comme la > logique est une morale de la pens?e - Jean Piaget > > > > Please, if possible, don't send me MS Word or > PowerPoint attachments > > Why? See: > http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National > Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > La morale est une logique de l'action comme la logique > est une morale de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or > PowerPoint attachments > Why? See: > http://www.gnu.org/philosophy/no-word-attachments.html > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From apusarkar at gmail.com Tue May 5 05:53:57 2009 From: apusarkar at gmail.com (Apu Sarkar) Date: Tue, 5 May 2009 09:23:57 +0530 Subject: [Pw_forum] Quantum Espresso Message-ID: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> Dear All, I am new to Quantum Espresso. I have configured and compiled the code witout any error in my laptop (1GB ram) with Linux Fedora core 9. Now when I am going to test the code with ./check-pw.x.j I am getting the message checking usppl_coulmb...... operating system error: cannot allocate memory Out of memory How can I get rid of this message and use the code? Thanking you and looking forward for your kind help. With best regards, Dr. A. Sarkar Materials Group Bhabha Atomic research Centre Mumbai 400085 India From degironc at sissa.it Tue May 5 08:26:01 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 05 May 2009 08:26:01 +0200 Subject: [Pw_forum] Virtual crystal approximation? In-Reply-To: References: <61853.20015.qm@web15203.mail.cnb.yahoo.com> Message-ID: <49FFDBF9.20807@sissa.it> Dear Wangwei, concerning the virtual.x code mentioned by Lorenzo Paulatto, please take notice that very recently (after your request to the forum) an updated version have been uploaded in the CVS repository incorporating the changes introduced by Jivtesh Garg and Nicola Marzari to fix the code misbehavior in the (common) case where the two pseudopotentials to be mixed have different mesh definitions. Please update your virtual.f90 version and report any problem or misbehavior you may notice. best regards, Stefano de Gironcoli - SISSA and DEMOCRITOS Lorenzo Paulatto wrote: > In data 29 aprile 2009 alle ore 12:11:04, wangwei > ha scritto: > >> Whether or not the PWSCF package can calculate by virtual crystal >> approximation for doped material? >> > > Dear Wangwei, > there is the virtual.x utility in the upftools directory of QE > distribution that can be used to generate virtual pseudopotential, it's > not very tested nor very mantained, as it is not used very often, but as > far as I know it works. Please have a look at the mailing list archives > here > > where the usage of virtual.x has been discussed extensively. > > regards > > > From sclauzer at sissa.it Tue May 5 08:43:00 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 05 May 2009 08:43:00 +0200 Subject: [Pw_forum] Quantum Espresso In-Reply-To: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> References: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> Message-ID: <49FFDFF4.3040103@sissa.it> Apu Sarkar wrote: > Dear All, > I am new to Quantum Espresso. > I have configured and compiled the code witout any error in my laptop > (1GB ram) with Linux Fedora core 9. > > Now when I am going to test the code with > ./check-pw.x.j > > I am getting the message > checking usppl_coulmb...... operating system error: cannot allocate memory > Out of memory > > How can I get rid of this message and use the code? Try ulimit -a and if the stack size is too small ulimit -s unlimited ./check-pw.x.j HTH GS > > Thanking you and looking forward for your kind help. > > With best regards, > > Dr. A. Sarkar > Materials Group > Bhabha Atomic research Centre > Mumbai 400085 > India > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Tue May 5 08:49:31 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 05 May 2009 08:49:31 +0200 Subject: [Pw_forum] Quantum Espresso In-Reply-To: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> References: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> Message-ID: <49FFE17B.1030305@sissa.it> BTW, we know already that this forum is about Quantum Espresso... ;-) It would be more useful if all posts which start a discussion have a less broad title... GS Apu Sarkar wrote: > Dear All, > I am new to Quantum Espresso. > I have configured and compiled the code witout any error in my laptop > (1GB ram) with Linux Fedora core 9. > > Now when I am going to test the code with > ./check-pw.x.j > > I am getting the message > checking usppl_coulmb...... operating system error: cannot allocate memory > Out of memory > > How can I get rid of this message and use the code? > > Thanking you and looking forward for your kind help. > > With best regards, > > Dr. A. Sarkar > Materials Group > Bhabha Atomic research Centre > Mumbai 400085 > India > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From giannozz at democritos.it Tue May 5 09:01:23 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 05 May 2009 09:01:23 +0200 Subject: [Pw_forum] Quantum Espresso In-Reply-To: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> References: <6293cd8f0905042053l78dbdb4dxb86a1a72c6ca7b6b@mail.gmail.com> Message-ID: <49FFE443.1030205@democritos.it> Apu Sarkar wrote: > I have configured and compiled the code the cvs version, or the stable version? > checking usppl_coulmb...... operating system error: cannot allocate memory > Out of memory > > How can I get rid of this message and use the code? is "uspp1_coulomb" the only test that shows a problem? if so, you can safely use the code, except the rather peculiar case of coulomb pseudopotentials (i.e. no pseudopotential but a true coulomb 1/r potential, typically used for high-precision calculations involving H atoms). The case of coulomb pseudopotentials turned out to contain some bugs, but it is hard to say whether they are involved in your problem without more information Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy From sclauzer at sissa.it Tue May 5 09:26:34 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 05 May 2009 09:26:34 +0200 Subject: [Pw_forum] dos In-Reply-To: <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> Message-ID: <49FFEA2A.7020908@sissa.it> Stefano Baroni wrote: > Jonas: I am afraid that what you want to obtain does not make much > sense. The angular character of any molecular (or Bloch) state depends > on the origin of your reference system. What is "s-like" with respect to > one point has an infinite number of angular components with respect to a > different point. Are you sure you want to add "what is s-like in Iowa" > with "what is s-like in Italy" ??? ;-) > Stefano I agree with what you say, Stefano, about taking the expectation value of the angular momentum in a periodic system, but summing the PDOS of selected atoms can lead to meaningful and sometimes very useful results, in some cases. I usually see the PDOS as an indicator of how much the Bloch states retain a kind of "atomic character" in the solid (so that in this case summing PDOS of equivalent atoms is the more meaningful thing, as Lex pointed out), but also as a way to identify the symmetry of states. If you place a molecule, a nanowire or maybe even a surface with slab geometry in a wise manner in your (super-)cell you can get many information from the PDOS (summed or not). Obviously, it does not make much sense to sum the s,p,d PDOS of all atoms in the system in general (but in some cases it may), so that's why you cannot find it directly in the output from projwfc. It's your task to select over which atoms you can do a sum (or an average) of the PDOS (and also placing the atoms in a smart way, first). I think you can write a script in awk/perl/whatever, or even a small fortran program, which reads PDOS from individual files associated to the selected atoms and sums the corresponding rows. Gabriele > > On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > >> >> I just successfully calculated DOS and PDOS of a very large system. >> One thing that confuses me is the presence of s, p components to PDOS >> for every atom, but not summed (or averaged) s and p contribution of >> all system. I would be happy to find in my output total DOS, PDOS s >> character and PDOS p character. Is there a way to get that instead of >> contribution per every atom? Or how do I proceed to get s and p PDOS >> for all system by manually processing all of my 76 atoms PDOS outputs?.. >> >> thanks >> >> Jonas Baltrusaitis >> University of Iowa >> >> >> --- On Mon, 5/4/09, Prasenjit Ghosh > > wrote: >> >>> From: Prasenjit Ghosh >> > >>> Subject: Re: [Pw_forum] dos >>> To: jasius_1 at yahoo.com , "PWSCF Forum" >>> > >>> Date: Monday, May 4, 2009, 12:04 AM >>> Hi Jonas, >>> >>> For DOS calculations one needs to do an integration over >>> the kpts in the >>> Brillouin zone. So you need a very fine k-point mesh than >>> necessary for a >>> scf calculation. Moreover you need a large number of bands. >>> Doing an scf >>> calculation with such a fine k-point mesh and so many bands >>> is very >>> expensive. So you should first do an scf calculation with >>> the min. no. of >>> k-points and bands and then do a nscf calculation with a >>> finer k-point and >>> larger no. of bands. In the nscf step, the code does an >>> interpolation for >>> the extra bands and k-points and not a self-consistent >>> calculation. This >>> saves computational time. >>> >>> Hope this helps. >>> >>> Prasenjit. >>> >>> 2009/5/3 Jonas Baltrusaitis >> > >>> >>>> >>>> Prasenjit, >>>> >>>> could you explain why both scf followed by nscf >>> calculations are needed for >>>> DOS calculations, as per example 8? If only scf >>> density is needed for >>>> further projection, why not to calculate only one? >>>> >>>> Jonas >>>> >>>> >>>> --- On Sun, 5/3/09, Prasenjit Ghosh >>> > wrote: >>>> >>>>> From: Prasenjit Ghosh >>> > >>>>> Subject: Re: [Pw_forum] dos >>>>> To: "PWSCF Forum" >>> > >>>>> Date: Sunday, May 3, 2009, 2:00 PM >>>>> dear all user >>>>>> >>>>> >>>>> Dear Yaldaa, >>>>> >>>>> >>>>>> How can I project dos to molecular orbital? >>>>>> >>>>> >>>>> There is a post-processing tool called projwfc.x >>> which you >>>>> can use to >>>>> project dos to molecular orbital. >>>>> >>>>> what is the importance of nbnd in calculations >>> for dos? >>>>>> >>>>> e >>>>> "nbnd" tells the code the no. of >>> Kohn-Sham >>>>> states it has to >>>>> compute.....While computing the dos, one wants to >>> plot the >>>>> dos for both the >>>>> occupied & unoccupied manifold, the no. of >>> energy >>>>> levels is read in by the >>>>> dos.x from the output of your scf/nscf run. So >>> while doing >>>>> an scf/ nscf >>>>> calculations you can control the total no. of >>> states with >>>>> the variable >>>>> "nbnd". >>>>> Depending on the type of occupation scheme you >>> choose the >>>>> value of nbnd is >>>>> nelec/2 (fixed occupation) or nelec/2+some extra >>> bands >>>>> (when you use >>>>> smearing). If you want to plot the dos, usually >>> you should >>>>> add some extra >>>>> levels using nbnd & do the scf or nscf >>> calculations to >>>>> get the unoccupied >>>>> manifold. >>>>> >>>>> and for plottingdos or pdos, if the fermi energy >>> is assumed >>>>> to be zero >>>>>> automatically by program? >>>>>> >>>>> >>>>> No, you have to do the alignment yourself >>>>> >>>>> With regards, >>>>> >>>>> Prasenjit. >>>>> >>>>> >>>>> >>>>> -- >>>>> PRASENJIT GHOSH, >>>>> POST-DOC, >>>>> ROOM NO: 265, MAIN BUILDING, >>>>> CM SECTION, ICTP, >>>>> STRADA COSTERIA 11, >>>>> TRIESTE, 34104, >>>>> ITALY >>>>> PHONE: +39 040 2240 369 (O) >>>>> +39 3807528672 (M) >>>>> _______________________________________________ >>>>> Pw_forum mailing list >>>>> Pw_forum at pwscf.org >>>>> >>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>>> >>>> >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>> >>> >>> >>> -- >>> PRASENJIT GHOSH, >>> POST-DOC, >>> ROOM NO: 265, MAIN BUILDING, >>> CM SECTION, ICTP, >>> STRADA COSTERIA 11, >>> TRIESTE, 34104, >>> ITALY >>> PHONE: +39 040 2240 369 (O) >>> +39 3807528672 (M) >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une morale de > la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From yaldachem at gmail.com Tue May 5 10:16:09 2009 From: yaldachem at gmail.com (yaldaa kh) Date: Tue, 5 May 2009 01:16:09 -0700 Subject: [Pw_forum] dos Message-ID: <47431d9f0905050116s475cde6cg13135f8681dcc00e@mail.gmail.com> Dear all please explain for plotting dos/pdos of different systems (in one panel) fermi energy is set zero, while the values of their fermi energy are different. now, what will be the scale of horizontal axis? thanks yalda student -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090505/b864caed/attachment.htm From paulatto at sissa.it Tue May 5 10:54:16 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 05 May 2009 10:54:16 +0200 Subject: [Pw_forum] dos In-Reply-To: <47431d9f0905050116s475cde6cg13135f8681dcc00e@mail.gmail.com> References: <47431d9f0905050116s475cde6cg13135f8681dcc00e@mail.gmail.com> Message-ID: In data 05 maggio 2009 alle ore 10:16:09, yaldaa kh ha scritto: > Dear all > please explain > for plotting dos/pdos of different systems (in one panel) fermi energy is > set zero, while the values of their fermi energy are different. > now, what will be the scale of horizontal axis? Dear Yaldaa, I'm not sure I'm giving you the answer you wanted, yet this is what I can say: pdos/dos output file are NOT scaled to bring the Fermi energy to zero. E.g. if you are using gnuplot and you want to have Ef=0 in the plot you can use a command like this: plot 'dos.dat' u ($1-Ef):2 wher Ef is (guess what?) the Fermi energy you can get from the output of the scf calculation. The scale of energy (horizontal axes) is in eV. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From kvvamsi20 at gmail.com Tue May 5 11:04:00 2009 From: kvvamsi20 at gmail.com (K V Vamsi) Date: Tue, 5 May 2009 14:34:00 +0530 Subject: [Pw_forum] bonding charge density. Message-ID: Hi all, I am new to quantum espresso. what is bonding charge density and how it is calculated exactly?. I have seen some tutorials, but program called chdens.x is required for that it is not present in my package here. how to go about it? can u help me..... -- Regards K V Vamsi Materials Engineering Deformation mechanisms and modelling group IISc, Bangalore - 560 012 Mob : 09886818221 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090505/1cf66ca3/attachment.htm From paulatto at sissa.it Tue May 5 11:11:25 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 05 May 2009 11:11:25 +0200 Subject: [Pw_forum] bonding charge density. In-Reply-To: References: Message-ID: In data 05 maggio 2009 alle ore 11:04:00, K V Vamsi ha scritto: > I am new to quantum espresso. Welcome KV, > what is bonding charge density and how it is calculated exactly?. QE uses a plane-wave basis set: in this formalism it doesn't make much sense to define a bonding charge density, yet it can help intuition. > I have seen some tutorials, but program called chdens.x is required for > that it is not present in my package here. They are probably old tutorial, since a few years ago chdens.x has been unified with pp.x, hence you can use pp whenever chdens is required, with minimal modifications to the input files. > how to go about it? can u help me..... Having a look at the examples is a good point to start, if you have some specific problem feel free to ask, but please try and search the mailing list archives before. Good work, best regards. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From sclauzer at sissa.it Tue May 5 11:18:04 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 05 May 2009 11:18:04 +0200 Subject: [Pw_forum] bonding charge density. In-Reply-To: References: Message-ID: <4A00044C.4030602@sissa.it> K V Vamsi wrote: > Hi all, > > I am new to quantum espresso. > what is bonding charge density and how it is calculated exactly?. Why do you want to calculate it if you don't know what it is exactly? > > I have seen some tutorials, but program called chdens.x is required for > that it is not present in my package here. > how to go about it? can u help me..... I think chdens.x was present in the old distributions of QE, now that part of the code should be included in pp.x (you can find the code in PP/ directory and the manual in Doc/INPUT_PP.txt). It can be used to extract and plot several quantities from the results of pw.x calculations (charge densities, KS eigenstates, spin polarization...) HTH GS > > -- > Regards > > K V Vamsi > Materials Engineering > Deformation mechanisms and modelling group > IISc, Bangalore - 560 012 > Mob : 09886818221 > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From jasius_1 at yahoo.com Tue May 5 15:02:09 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 5 May 2009 06:02:09 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <49FFEA2A.7020908@sissa.it> Message-ID: <469132.62825.qm@web54401.mail.yahoo.com> Gabriele, you mean physically sum up all, say, C2s PDOS? There is no averaging or similar task involved? --- On Tue, 5/5/09, Gabriele Sclauzero wrote: > From: Gabriele Sclauzero > Subject: Re: [Pw_forum] dos > To: "PWSCF Forum" > Date: Tuesday, May 5, 2009, 12:26 AM > Stefano Baroni wrote: > > Jonas: I am afraid that what you want to obtain does > not make much > > sense. The angular character of any molecular (or > Bloch) state depends > > on the origin of your reference system. What is > "s-like" with respect to > > one point has an infinite number of angular components > with respect to a > > different point. Are you sure you want to add > "what is s-like in Iowa" > > with "what is s-like in Italy" ??? ;-) > > Stefano > > I agree with what you say, Stefano, about taking the > expectation value of the angular > momentum in a periodic system, but summing the PDOS of > selected atoms can lead to > meaningful and sometimes very useful results, in some > cases. > I usually see the PDOS as an indicator of how much the > Bloch states retain a kind of > "atomic character" in the solid (so that in this > case summing PDOS of equivalent atoms is > the more meaningful thing, as Lex pointed out), but also as > a way to identify the symmetry > of states. > If you place a molecule, a nanowire or maybe even a surface > with slab geometry in a wise > manner in your (super-)cell you can get many information > from the PDOS (summed or not). > Obviously, it does not make much sense to sum the s,p,d > PDOS of all atoms in the system in > general (but in some cases it may), so that's why you > cannot find it directly in the > output from projwfc. > It's your task to select over which atoms you can do a > sum (or an average) of the PDOS > (and also placing the atoms in a smart way, first). I think > you can write a script in > awk/perl/whatever, or even a small fortran program, which > reads PDOS from individual files > associated to the selected atoms and sums the corresponding > rows. > > Gabriele > > > > > On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > > > >> > >> I just successfully calculated DOS and PDOS of a > very large system. > >> One thing that confuses me is the presence of s, p > components to PDOS > >> for every atom, but not summed (or averaged) s and > p contribution of > >> all system. I would be happy to find in my output > total DOS, PDOS s > >> character and PDOS p character. Is there a way to > get that instead of > >> contribution per every atom? Or how do I proceed > to get s and p PDOS > >> for all system by manually processing all of my 76 > atoms PDOS outputs?.. > >> > >> thanks > >> > >> Jonas Baltrusaitis > >> University of Iowa > >> > >> > >> --- On Mon, 5/4/09, Prasenjit Ghosh > >> > wrote: > >> > >>> From: Prasenjit Ghosh > >>> > > >>> Subject: Re: [Pw_forum] dos > >>> To: jasius_1 at yahoo.com > , "PWSCF Forum" > >>> > > >>> Date: Monday, May 4, 2009, 12:04 AM > >>> Hi Jonas, > >>> > >>> For DOS calculations one needs to do an > integration over > >>> the kpts in the > >>> Brillouin zone. So you need a very fine > k-point mesh than > >>> necessary for a > >>> scf calculation. Moreover you need a large > number of bands. > >>> Doing an scf > >>> calculation with such a fine k-point mesh and > so many bands > >>> is very > >>> expensive. So you should first do an scf > calculation with > >>> the min. no. of > >>> k-points and bands and then do a nscf > calculation with a > >>> finer k-point and > >>> larger no. of bands. In the nscf step, the > code does an > >>> interpolation for > >>> the extra bands and k-points and not a > self-consistent > >>> calculation. This > >>> saves computational time. > >>> > >>> Hope this helps. > >>> > >>> Prasenjit. > >>> > >>> 2009/5/3 Jonas Baltrusaitis > >>> > > >>> > >>>> > >>>> Prasenjit, > >>>> > >>>> could you explain why both scf followed by > nscf > >>> calculations are needed for > >>>> DOS calculations, as per example 8? If > only scf > >>> density is needed for > >>>> further projection, why not to calculate > only one? > >>>> > >>>> Jonas > >>>> > >>>> > >>>> --- On Sun, 5/3/09, Prasenjit Ghosh > >>> > wrote: > >>>> > >>>>> From: Prasenjit Ghosh > >>> > > >>>>> Subject: Re: [Pw_forum] dos > >>>>> To: "PWSCF Forum" > >>> > > >>>>> Date: Sunday, May 3, 2009, 2:00 PM > >>>>> dear all user > >>>>>> > >>>>> > >>>>> Dear Yaldaa, > >>>>> > >>>>> > >>>>>> How can I project dos to molecular > orbital? > >>>>>> > >>>>> > >>>>> There is a post-processing tool called > projwfc.x > >>> which you > >>>>> can use to > >>>>> project dos to molecular orbital. > >>>>> > >>>>> what is the importance of nbnd in > calculations > >>> for dos? > >>>>>> > >>>>> e > >>>>> "nbnd" tells the code the > no. of > >>> Kohn-Sham > >>>>> states it has to > >>>>> compute.....While computing the dos, > one wants to > >>> plot the > >>>>> dos for both the > >>>>> occupied & unoccupied manifold, > the no. of > >>> energy > >>>>> levels is read in by the > >>>>> dos.x from the output of your scf/nscf > run. So > >>> while doing > >>>>> an scf/ nscf > >>>>> calculations you can control the total > no. of > >>> states with > >>>>> the variable > >>>>> "nbnd". > >>>>> Depending on the type of occupation > scheme you > >>> choose the > >>>>> value of nbnd is > >>>>> nelec/2 (fixed occupation) or > nelec/2+some extra > >>> bands > >>>>> (when you use > >>>>> smearing). If you want to plot the > dos, usually > >>> you should > >>>>> add some extra > >>>>> levels using nbnd & do the scf or > nscf > >>> calculations to > >>>>> get the unoccupied > >>>>> manifold. > >>>>> > >>>>> and for plottingdos or pdos, if the > fermi energy > >>> is assumed > >>>>> to be zero > >>>>>> automatically by program? > >>>>>> > >>>>> > >>>>> No, you have to do the alignment > yourself > >>>>> > >>>>> With regards, > >>>>> > >>>>> Prasenjit. > >>>>> > >>>>> > >>>>> > >>>>> -- > >>>>> PRASENJIT GHOSH, > >>>>> POST-DOC, > >>>>> ROOM NO: 265, MAIN BUILDING, > >>>>> CM SECTION, ICTP, > >>>>> STRADA COSTERIA 11, > >>>>> TRIESTE, 34104, > >>>>> ITALY > >>>>> PHONE: +39 040 2240 369 (O) > >>>>> +39 3807528672 (M) > >>>>> > _______________________________________________ > >>>>> Pw_forum mailing list > >>>>> Pw_forum at pwscf.org > > >>>>> > >>> > http://www.democritos.it/mailman/listinfo/pw_forum > >>>> > >>>> > >>>> > >>>> > _______________________________________________ > >>>> Pw_forum mailing list > >>>> Pw_forum at pwscf.org > > >>>> > http://www.democritos.it/mailman/listinfo/pw_forum > >>>> > >>> > >>> > >>> > >>> -- > >>> PRASENJIT GHOSH, > >>> POST-DOC, > >>> ROOM NO: 265, MAIN BUILDING, > >>> CM SECTION, ICTP, > >>> STRADA COSTERIA 11, > >>> TRIESTE, 34104, > >>> ITALY > >>> PHONE: +39 040 2240 369 (O) > >>> +39 3807528672 (M) > >> > >> > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > > >> http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > > Stefano Baroni - SISSA & DEMOCRITOS National > Simulation Center - Trieste > > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni > (skype) > > > > La morale est une logique de l'action comme la > logique est une morale de > > la pens?e - Jean Piaget > > > > Please, if possible, don't send me MS Word or > PowerPoint attachments > > Why? See: > http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From sclauzer at sissa.it Tue May 5 15:42:10 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 05 May 2009 15:42:10 +0200 Subject: [Pw_forum] dos In-Reply-To: <469132.62825.qm@web54401.mail.yahoo.com> References: <469132.62825.qm@web54401.mail.yahoo.com> Message-ID: <4A004232.6070503@sissa.it> Jonas Baltrusaitis wrote: > Gabriele, > > you mean physically sum up all, say, C2s PDOS? There is no averaging or similar task involved? It depends what you want to look at... for instance, if I want to find the energy position of the \pi states of a CO molecule (aligned along z) I could sum up the s and p_z PDOS of C and O atoms. Since the units of PDOS are electrons/ev (apart from some spilling charge due to incompleteness of atomic basis in the solid), I could integrate the summed PDOS and find that each \pi peak corresponds to an integer number of electrons (which is given by the degeneracy of the state). Averaging does not make much sense in this case, since C and O are obviously not equivalent. At the moment I cannot figure out a case where it could have. Also averaging PDOS of equivalent atoms is not useful, since PDOS on all equivalent atoms should be equal. I don't know what else you were meaning by "averaging", sorry. GS > > > --- On Tue, 5/5/09, Gabriele Sclauzero wrote: > >> From: Gabriele Sclauzero >> Subject: Re: [Pw_forum] dos >> To: "PWSCF Forum" >> Date: Tuesday, May 5, 2009, 12:26 AM >> Stefano Baroni wrote: >>> Jonas: I am afraid that what you want to obtain does >> not make much >>> sense. The angular character of any molecular (or >> Bloch) state depends >>> on the origin of your reference system. What is >> "s-like" with respect to >>> one point has an infinite number of angular components >> with respect to a >>> different point. Are you sure you want to add >> "what is s-like in Iowa" >>> with "what is s-like in Italy" ??? ;-) >>> Stefano >> I agree with what you say, Stefano, about taking the >> expectation value of the angular >> momentum in a periodic system, but summing the PDOS of >> selected atoms can lead to >> meaningful and sometimes very useful results, in some >> cases. >> I usually see the PDOS as an indicator of how much the >> Bloch states retain a kind of >> "atomic character" in the solid (so that in this >> case summing PDOS of equivalent atoms is >> the more meaningful thing, as Lex pointed out), but also as >> a way to identify the symmetry >> of states. >> If you place a molecule, a nanowire or maybe even a surface >> with slab geometry in a wise >> manner in your (super-)cell you can get many information >> from the PDOS (summed or not). >> Obviously, it does not make much sense to sum the s,p,d >> PDOS of all atoms in the system in >> general (but in some cases it may), so that's why you >> cannot find it directly in the >> output from projwfc. >> It's your task to select over which atoms you can do a >> sum (or an average) of the PDOS >> (and also placing the atoms in a smart way, first). I think >> you can write a script in >> awk/perl/whatever, or even a small fortran program, which >> reads PDOS from individual files >> associated to the selected atoms and sums the corresponding >> rows. >> >> Gabriele >> >>> On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: >>> >>>> I just successfully calculated DOS and PDOS of a >> very large system. >>>> One thing that confuses me is the presence of s, p >> components to PDOS >>>> for every atom, but not summed (or averaged) s and >> p contribution of >>>> all system. I would be happy to find in my output >> total DOS, PDOS s >>>> character and PDOS p character. Is there a way to >> get that instead of >>>> contribution per every atom? Or how do I proceed >> to get s and p PDOS >>>> for all system by manually processing all of my 76 >> atoms PDOS outputs?.. >>>> thanks >>>> >>>> Jonas Baltrusaitis >>>> University of Iowa -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From akohlmey at cmm.chem.upenn.edu Tue May 5 16:02:03 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 05 May 2009 10:02:03 -0400 Subject: [Pw_forum] dos In-Reply-To: <4A004232.6070503@sissa.it> References: <469132.62825.qm@web54401.mail.yahoo.com> <4A004232.6070503@sissa.it> Message-ID: <1241532123.2960.17.camel@zero> On Tue, 2009-05-05 at 15:42 +0200, Gabriele Sclauzero wrote: > Also averaging PDOS of equivalent atoms is not useful, since PDOS on all equivalent atoms > should be equal. > I don't know what else you were meaning by "averaging", sorry. it may be too obvious, but i would like to point out that the difference between summing up and averaging is just a scaling factor. taking into account all the concerns about comparing apples with oranges, the only scenario where an average would make sense is to compare different size supercells with otherwise identical structures. cheers, axel. > GS > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From kvvamsi20 at gmail.com Tue May 5 16:53:12 2009 From: kvvamsi20 at gmail.com (K V Vamsi) Date: Tue, 5 May 2009 20:23:12 +0530 Subject: [Pw_forum] Pw_forum Digest, Vol 23, Issue 11 In-Reply-To: References: Message-ID: Hi, > Thank you for your response. Till now i worked with charge density plots. > Now i want to see the specific effect of an element on an alloy system. So i > want to know what king of difference it make in bonding to the alloy system. > so, I am interested in plotting bonding charge density plots ( i have come > to know this for literature available) which give information regarding > bonding. > I dont know how to plot these bonding charge density plots using pp.x can u help me.... > > ------------------------------ > > Message: 6 > Date: Tue, 05 May 2009 11:18:04 +0200 > From: Gabriele Sclauzero > Subject: Re: [Pw_forum] bonding charge density. > To: PWSCF Forum > Message-ID: <4A00044C.4030602 at sissa.it> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > K V Vamsi wrote: > > Hi all, > > > > I am new to quantum espresso. > > what is bonding charge density and how it is calculated exactly?. > > Why do you want to calculate it if you don't know what it is exactly? > > > > > I have seen some tutorials, but program called chdens.x is required for > > that it is not present in my package here. > > how to go about it? can u help me..... > > I think chdens.x was present in the old distributions of QE, now that part > of the code > should be included in pp.x (you can find the code in PP/ directory and the > manual in > Doc/INPUT_PP.txt). It can be used to extract and plot several quantities > from the results > of pw.x calculations (charge densities, KS eigenstates, spin > polarization...) > > > HTH > > GS > > > > > > -- > > Regards > > > > K V Vamsi > > Materials Engineering > > Deformation mechanisms and modelling group > > IISc, Bangalore - 560 012 > > Mob : 09886818221 > > -- Regards K V Vamsi Materials Engineering Deformation mechanisms and modelling group IISc, Bangalore - 560 012 Mob : 09886818221 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090505/6430acca/attachment.htm From paulatto at sissa.it Tue May 5 17:07:39 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 05 May 2009 17:07:39 +0200 Subject: [Pw_forum] Pw_forum Digest, Vol 23, Issue 11 In-Reply-To: References: Message-ID: In data 05 maggio 2009 alle ore 16:53:12, K V Vamsi ha scritto: > I dont know how to plot these bonding charge density plots using pp.x > can u help me.... Deal KV, I assume you have been asked to plot bonding charge densities without having been explained what they actually are. Note that it is not possible to precisely *define* bonding charge density in Quantum-ESPRESSO. If you have found a tutorial stating otherwise, send it (or a link to it) to the mailing list, so we can have a better idea of what you *need* to do, instead of what you want to do, as the latter is unclear. In case you are referring to this , it is very similar to example05 included in the example directory, please have a look at that. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From baroni at sissa.it Tue May 5 20:33:54 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 5 May 2009 20:33:54 +0200 Subject: [Pw_forum] dos In-Reply-To: <49FFEA2A.7020908@sissa.it> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> Message-ID: <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> I am still rather skeptical. If you apply your reasoning to the free electron gas, it seems to me that you would obtain rather weird (to say the least) results. I may be wong, and finally the choice on how to analyze your data is yours. SB On May 5, 2009, at 9:26 AM, Gabriele Sclauzero wrote: > > Stefano Baroni wrote: >> Jonas: I am afraid that what you want to obtain does not make much >> sense. The angular character of any molecular (or Bloch) state >> depends >> on the origin of your reference system. What is "s-like" with >> respect to >> one point has an infinite number of angular components with respect >> to a >> different point. Are you sure you want to add "what is s-like in >> Iowa" >> with "what is s-like in Italy" ??? ;-) >> Stefano > > I agree with what you say, Stefano, about taking the expectation > value of the angular > momentum in a periodic system, but summing the PDOS of selected > atoms can lead to > meaningful and sometimes very useful results, in some cases. > I usually see the PDOS as an indicator of how much the Bloch states > retain a kind of > "atomic character" in the solid (so that in this case summing PDOS > of equivalent atoms is > the more meaningful thing, as Lex pointed out), but also as a way to > identify the symmetry > of states. > If you place a molecule, a nanowire or maybe even a surface with > slab geometry in a wise > manner in your (super-)cell you can get many information from the > PDOS (summed or not). > Obviously, it does not make much sense to sum the s,p,d PDOS of all > atoms in the system in > general (but in some cases it may), so that's why you cannot find it > directly in the > output from projwfc. > It's your task to select over which atoms you can do a sum (or an > average) of the PDOS > (and also placing the atoms in a smart way, first). I think you can > write a script in > awk/perl/whatever, or even a small fortran program, which reads PDOS > from individual files > associated to the selected atoms and sums the corresponding rows. > > Gabriele > >> >> On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: >> >>> >>> I just successfully calculated DOS and PDOS of a very large system. >>> One thing that confuses me is the presence of s, p components to >>> PDOS >>> for every atom, but not summed (or averaged) s and p contribution of >>> all system. I would be happy to find in my output total DOS, PDOS s >>> character and PDOS p character. Is there a way to get that instead >>> of >>> contribution per every atom? Or how do I proceed to get s and p PDOS >>> for all system by manually processing all of my 76 atoms PDOS >>> outputs?.. >>> >>> thanks >>> >>> Jonas Baltrusaitis >>> University of Iowa >>> --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090505/79174d0c/attachment-0001.htm From kvvamsi20 at gmail.com Tue May 5 22:06:51 2009 From: kvvamsi20 at gmail.com (K V Vamsi) Date: Wed, 6 May 2009 01:36:51 +0530 Subject: [Pw_forum] charge density plots. Message-ID: Thank u, The following is the link through which i have come to know about the charge density difference plots. > But to plot this i dont have the chdens.x program in my package as its specified in the exercise 4. I saw that can this charge density difference can be plotted by pp.x.? presently i am working on Ni-Al-Ti system. I want to see the effect of Ti in Ni-Al system. I thought these plots could be of some help. -- Regards K V Vamsi Materials Engineering Deformation mechanisms and modelling group IISc, Bangalore - 560 012 Mob : 09886818221 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090506/40e15e34/attachment.htm From jasius_1 at yahoo.com Wed May 6 02:30:51 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 5 May 2009 17:30:51 -0700 (PDT) Subject: [Pw_forum] submits jobs to node already in use In-Reply-To: <627e0ffa0905040004j72f19d2fk7f8befa300ad06be@mail.gmail.com> Message-ID: <797670.7754.qm@web54402.mail.yahoo.com> Funny story, if I submit any job with mpiexec -n 4 -machinefile /../pw.x -rmpool 0 -nodes 1 -procs 4 via PBS to a 4 core processor I can't see the exact number of cores running, e.g. with qstat -n I see only compute-0-1/0, instead of usual compute-0-1/0, compute-0-1/1, compute-0-1/2, compute-0-1/4. It gets even worse if I want to submit any other pwscf job to my 16 node cluster. It get's submitted to exactly the same node, e.g. I run 8 processes on a 4 core node and obviously no performance I inquired at Rocks forum, they claim it must be related with how Q-E submits jobs. Has anybody seen that before? Jonas Baltrusaitis University of Iowa From akohlmey at cmm.chem.upenn.edu Wed May 6 05:29:49 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 05 May 2009 23:29:49 -0400 Subject: [Pw_forum] submits jobs to node already in use In-Reply-To: <797670.7754.qm@web54402.mail.yahoo.com> References: <797670.7754.qm@web54402.mail.yahoo.com> Message-ID: <1241580589.10739.4.camel@zero> On Tue, 2009-05-05 at 17:30 -0700, Jonas Baltrusaitis wrote: > Funny story, if I submit any job with mpiexec -n 4 -machinefile where is the machinefile?? > /../pw.x -rmpool 0 -nodes 1 -procs 4 via PBS to a 4 core processor I this must be a specialty of how your machine is set up. if you submit to a PBS queue, you don't use mpiexec, but rather use qsub. mpiexec (or mpirun) is being called from within the job script that you submit to qsub. > can't see the exact number of cores running, e.g. with qstat -n I see > only compute-0-1/0, instead of usual compute-0-1/0, compute-0-1/1, > compute-0-1/2, compute-0-1/4. again. this has nothing at all to do with Q-E. there is no code in Q-E that knows anything about any batch system (and it should not). > It gets even worse if I want to submit any other pwscf job to my 16 > node cluster. It get's submitted to exactly the same node, e.g. I run > 8 processes on a 4 core node and obviously no performance > > I inquired at Rocks forum, they claim it must be related with how Q-E > submits jobs. Has anybody seen that before? that is wrong advice. cheers, axel. > > Jonas Baltrusaitis > University of Iowa > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From paulatto at sissa.it Wed May 6 08:07:32 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 6 May 2009 08:07:32 +0200 (CEST) Subject: [Pw_forum] submits jobs to node already in use In-Reply-To: <797670.7754.qm@web54402.mail.yahoo.com> References: <797670.7754.qm@web54402.mail.yahoo.com> Message-ID: <40251.78.12.164.106.1241590052.squirrel@webmail.sissa.it> In data 06 maggio 2009 alle ore 02:30:51, Jonas Baltrusaitis ha scritto: It get's submitted to exactly the same node, e.g. I run 8 processes on a 4 core node and obviously no performance Last time this happened to I had to manually specify --mpipref among the mpirun options. I hope it helps, regards. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From kazempoor2000 at yahoo.com Wed May 6 08:23:40 2009 From: kazempoor2000 at yahoo.com (ali kazempoor) Date: Tue, 5 May 2009 23:23:40 -0700 (PDT) Subject: [Pw_forum] vc-relax Message-ID: <295302.54036.qm@web33104.mail.mud.yahoo.com> Dear All I want to study an O vacancy in 2*2*3 supercell for TiO2. I use from vc-relax to optimize the lattice constant either with atomic relaxation. But from one week ago, the pressure evolved around 0.00 and dident go to zero .for example (kbar) P= 0.07 (kbar) P= -0.45 (kbar) P= -0.23 (kbar) P= -0.25 (kbar) P= -0.34 (kbar) P= -0.40 (kbar) P= -0.36 (kbar) P= -0.14 (kbar) P= -0.57 (kbar) P= -0.41 (kbar) P= -0.45 (kbar) P= -0.83 (kbar) P= -0.45 (kbar) P= -0.37 (kbar) P= -1.58 (kbar) P= -1.00 (kbar) P= -0.71 (kbar) P= -0.73 (kbar) P= -0.61 (kbar) P= -0.58 (kbar) P= -0.39 (kbar) P= -0.89 (kbar) P= -0.58 (kbar) P= -0.33 (kbar) P= -0.37 (kbar) P= 0.59 (kbar) P= -0.42 (kbar) P= -0.61 (kbar) P= -1.22 (kbar) P= -1.68 (kbar) P= -0.23 (kbar) P= -0.20 (kbar) P= -0.17 (kbar) P= -0.39 (kbar) P= -0.29 (kbar) P= -0.76 (kbar) P= -0.30 (kbar) P= -0.24 what can I do for this problem? Here is input file: &control calculation = 'vc-relax' restart_mode='restart', prefix='rutile-71atom' tstress = .true. tprnfor = .true. pseudo_dir = './', outdir='/ptmp/javad/scratch/' nstep=1000, dt=100, / &system ibrav= 6 , celldm(1) =17.36,celldm(3)=0.9663585, nat= 71, ntyp= 2, ecutwfc =35.0,ecutrho=400 / &electrons diagonalization='' mixing_mode = 'plain' mixing_beta = 0.7 conv_thr = 1.0d-4 / &IONS / &CELL cell_dynamics = 'damp-w', press=00.0 / ATOMIC_SPECIES Ti 43.0 Ti.pz-sp-van.UPF O 16.0 O.pz-van_ak.UPF ATOMIC_POSITIONS crystal Ti 0.00000000 0.00000000 0.00000000 Ti 0.50000000 0.00000000 0.00000000 Ti 0.00000000 0.50000000 0.00000000 Ti 0.50000000 0.50000000 0.00000000 Ti 0.00000000 0.00000000 0.33333333 Ti 0.50000000 0.00000000 0.33333333 Ti 0.00000000 0.50000000 0.33333333 Ti 0.50000000 0.50000000 0.33333333 Ti 0.00000000 0.00000000 0.66666667 Ti 0.50000000 0.00000000 0.66666667 Ti 0.00000000 0.50000000 0.66666667 Ti 0.50000000 0.50000000 0.66666667 Ti 0.25000000 0.25000000 0.16666667 Ti 0.75000000 0.25000000 0.16666667 Ti 0.25000000 0.75000000 0.16666667 Ti 0.75000000 0.75000000 0.16666667 Ti 0.25000000 0.25000000 0.50000000 Ti 0.75000000 0.25000000 0.50000000 Ti 0.25000000 0.75000000 0.50000000 Ti 0.75000000 0.75000000 0.50000000 Ti 0.25000000 0.25000000 0.83333333 Ti 0.75000000 0.25000000 0.83333333 Ti 0.25000000 0.75000000 0.83333333 Ti 0.75000000 0.75000000 0.83333333 O 0.65265000 0.15265000 0.00000000 O 0.15265000 0.65265000 0.00000000 O 0.65265000 0.65265000 0.00000000 O 0.15265000 0.15265000 0.33333333 O 0.65265000 0.15265000 0.33333333 O 0.15265000 0.65265000 0.33333333 O 0.65265000 0.65265000 0.33333333 O 0.15265000 0.15265000 0.66666667 O 0.65265000 0.15265000 0.66666667 O 0.15265000 0.65265000 0.66666667 O 0.65265000 0.65265000 0.66666667 O 0.34735000 0.34735000 0.00000000 O 0.84735000 0.34735000 0.00000000 O 0.34735000 0.84735000 0.00000000 O 0.84735000 0.84735000 0.00000000 O 0.34735000 0.34735000 0.33333333 O 0.84735000 0.34735000 0.33333333 O 0.34735000 0.84735000 0.33333333 O 0.84735000 0.84735000 0.33333333 O 0.34735000 0.34735000 0.66666667 O 0.84735000 0.34735000 0.66666667 O 0.34735000 0.84735000 0.66666667 O 0.84735000 0.84735000 0.66666667 O 0.40265000 0.09735000 0.16666667 O 0.90265000 0.09735000 0.16666667 O 0.40265000 0.59735000 0.16666667 O 0.90265000 0.59735000 0.16666667 O 0.40265000 0.09735000 0.50000000 O 0.90265000 0.09735000 0.50000000 O 0.40265000 0.59735000 0.50000000 O 0.90265000 0.59735000 0.50000000 O 0.40265000 0.09735000 0.83333333 O 0.90265000 0.09735000 0.83333333 O 0.40265000 0.59735000 0.83333333 O 0.90265000 0.59735000 0.83333333 O 0.09735000 0.40265000 0.16666667 O 0.59735000 0.40265000 0.16666667 O 0.09735000 0.90265000 0.16666667 O 0.59735000 0.90265000 0.16666667 O 0.09735000 0.40265000 0.50000000 O 0.59735000 0.40265000 0.50000000 O 0.09735000 0.90265000 0.50000000 O 0.59735000 0.90265000 0.50000000 O 0.09735000 0.40265000 0.83333333 O 0.59735000 0.40265000 0.83333333 O 0.09735000 0.90265000 0.83333333 O 0.59735000 0.90265000 0.83333333 K_POINTS automatic 3 3 2 1 1 1 thanks alot Ali Kazempour,Isfahan University of Technology From degironc at sissa.it Wed May 6 08:39:03 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Wed, 06 May 2009 08:39:03 +0200 Subject: [Pw_forum] vc-relax In-Reply-To: <295302.54036.qm@web33104.mail.mud.yahoo.com> References: <295302.54036.qm@web33104.mail.mud.yahoo.com> Message-ID: <4A013087.5020800@sissa.it> have you tried to relax the internal coordinates before using vc-relax ? there is a coupling between internal and cell degrees of freedom and maybe what you see is just the effect of the internal degrees of freedom (that are many in your case) slowly (because of the damp dynamics algorithm) approaching their equilibrium values. have you tried the bfgs algorithm in vc-relax ? stefano de Gironcoli - SISSA and DEMOCRITOS ali kazempoor wrote: > Dear All > I want to study an O vacancy in 2*2*3 supercell for TiO2. I use from vc-relax to optimize the lattice constant either with atomic relaxation. But from one week ago, the pressure evolved around 0.00 and dident go to zero .for example > (kbar) P= 0.07 > (kbar) P= -0.45 > (kbar) P= -0.23 > (kbar) P= -0.25 > (kbar) P= -0.34 > (kbar) P= -0.40 > (kbar) P= -0.36 > (kbar) P= -0.14 > (kbar) P= -0.57 > (kbar) P= -0.41 > (kbar) P= -0.45 > (kbar) P= -0.83 > (kbar) P= -0.45 > (kbar) P= -0.37 > (kbar) P= -1.58 > (kbar) P= -1.00 > (kbar) P= -0.71 > (kbar) P= -0.73 > (kbar) P= -0.61 > (kbar) P= -0.58 > (kbar) P= -0.39 > (kbar) P= -0.89 > (kbar) P= -0.58 > (kbar) P= -0.33 > (kbar) P= -0.37 > (kbar) P= 0.59 > (kbar) P= -0.42 > (kbar) P= -0.61 > (kbar) P= -1.22 > (kbar) P= -1.68 > (kbar) P= -0.23 > (kbar) P= -0.20 > (kbar) P= -0.17 > (kbar) P= -0.39 > (kbar) P= -0.29 > (kbar) P= -0.76 > (kbar) P= -0.30 > (kbar) P= -0.24 > what can I do for this problem? > Here is input file: > > > > &control > calculation = 'vc-relax' > restart_mode='restart', > prefix='rutile-71atom' > tstress = .true. > tprnfor = .true. > pseudo_dir = './', > outdir='/ptmp/javad/scratch/' > nstep=1000, > dt=100, > / > &system > ibrav= 6 , celldm(1) =17.36,celldm(3)=0.9663585, nat= 71, ntyp= 2, > ecutwfc =35.0,ecutrho=400 > / > &electrons > diagonalization='' > mixing_mode = 'plain' > mixing_beta = 0.7 > conv_thr = 1.0d-4 > / > &IONS > / > &CELL > cell_dynamics = 'damp-w', > press=00.0 > / > ATOMIC_SPECIES > Ti 43.0 Ti.pz-sp-van.UPF > O 16.0 O.pz-van_ak.UPF > ATOMIC_POSITIONS crystal > Ti 0.00000000 0.00000000 0.00000000 > Ti 0.50000000 0.00000000 0.00000000 > Ti 0.00000000 0.50000000 0.00000000 > Ti 0.50000000 0.50000000 0.00000000 > Ti 0.00000000 0.00000000 0.33333333 > Ti 0.50000000 0.00000000 0.33333333 > Ti 0.00000000 0.50000000 0.33333333 > Ti 0.50000000 0.50000000 0.33333333 > Ti 0.00000000 0.00000000 0.66666667 > Ti 0.50000000 0.00000000 0.66666667 > Ti 0.00000000 0.50000000 0.66666667 > Ti 0.50000000 0.50000000 0.66666667 > Ti 0.25000000 0.25000000 0.16666667 > Ti 0.75000000 0.25000000 0.16666667 > Ti 0.25000000 0.75000000 0.16666667 > Ti 0.75000000 0.75000000 0.16666667 > Ti 0.25000000 0.25000000 0.50000000 > Ti 0.75000000 0.25000000 0.50000000 > Ti 0.25000000 0.75000000 0.50000000 > Ti 0.75000000 0.75000000 0.50000000 > Ti 0.25000000 0.25000000 0.83333333 > Ti 0.75000000 0.25000000 0.83333333 > Ti 0.25000000 0.75000000 0.83333333 > Ti 0.75000000 0.75000000 0.83333333 > O 0.65265000 0.15265000 0.00000000 > O 0.15265000 0.65265000 0.00000000 > O 0.65265000 0.65265000 0.00000000 > O 0.15265000 0.15265000 0.33333333 > O 0.65265000 0.15265000 0.33333333 > O 0.15265000 0.65265000 0.33333333 > O 0.65265000 0.65265000 0.33333333 > O 0.15265000 0.15265000 0.66666667 > O 0.65265000 0.15265000 0.66666667 > O 0.15265000 0.65265000 0.66666667 > O 0.65265000 0.65265000 0.66666667 > O 0.34735000 0.34735000 0.00000000 > O 0.84735000 0.34735000 0.00000000 > O 0.34735000 0.84735000 0.00000000 > O 0.84735000 0.84735000 0.00000000 > O 0.34735000 0.34735000 0.33333333 > O 0.84735000 0.34735000 0.33333333 > O 0.34735000 0.84735000 0.33333333 > O 0.84735000 0.84735000 0.33333333 > O 0.34735000 0.34735000 0.66666667 > O 0.84735000 0.34735000 0.66666667 > O 0.34735000 0.84735000 0.66666667 > O 0.84735000 0.84735000 0.66666667 > O 0.40265000 0.09735000 0.16666667 > O 0.90265000 0.09735000 0.16666667 > O 0.40265000 0.59735000 0.16666667 > O 0.90265000 0.59735000 0.16666667 > O 0.40265000 0.09735000 0.50000000 > O 0.90265000 0.09735000 0.50000000 > O 0.40265000 0.59735000 0.50000000 > O 0.90265000 0.59735000 0.50000000 > O 0.40265000 0.09735000 0.83333333 > O 0.90265000 0.09735000 0.83333333 > O 0.40265000 0.59735000 0.83333333 > O 0.90265000 0.59735000 0.83333333 > O 0.09735000 0.40265000 0.16666667 > O 0.59735000 0.40265000 0.16666667 > O 0.09735000 0.90265000 0.16666667 > O 0.59735000 0.90265000 0.16666667 > O 0.09735000 0.40265000 0.50000000 > O 0.59735000 0.40265000 0.50000000 > O 0.09735000 0.90265000 0.50000000 > O 0.59735000 0.90265000 0.50000000 > O 0.09735000 0.40265000 0.83333333 > O 0.59735000 0.40265000 0.83333333 > O 0.09735000 0.90265000 0.83333333 > O 0.59735000 0.90265000 0.83333333 > K_POINTS automatic > 3 3 2 1 1 1 > thanks alot > Ali Kazempour,Isfahan University of Technology > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From sclauzer at sissa.it Wed May 6 08:57:06 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Wed, 06 May 2009 08:57:06 +0200 Subject: [Pw_forum] charge density plots. In-Reply-To: References: Message-ID: <4A0134C2.7010103@sissa.it> Dear Vamsi, please try not to change the subject of this discussion too many times. The best way to continue a discussion in the forum is to reply to a previous email from the forum itself AND belonging to that discussion/topic, so that other users can follow the thread. Thanks. K V Vamsi wrote: > Thank u, > The following is the link through which i have come to know about the > charge density difference plots. > > > But to plot this i dont have the chdens.x program in my package as its > specified in the exercise 4. Again, the functionalities of chdens.x are now included in pp.x, please take few minutes to read the INPUT_PP file. > I saw that can this charge density difference can be plotted by pp.x.? Yes, of course, but you cannot do it in a single step, I think. First produce three charge density files feeding pp.x with something like: &inputpp outdir= "/scratch/" prefix= "prefix1", filplot= "file1", plot_num= 0 / and then &inputpp outdir= "/scratch/" prefix= "prefix2", filplot= "file2", plot_num= 0 / and then &inputpp outdir= "/scratch/" prefix= "prefix3", filplot= "file3", plot_num= 0 / Then run again pp.x with something like &inputpp / &plot nfile=3 filepp(1)='file1', weight(1)=1.d0 filepp(2)='file2', weight(2)=-1.d0 filepp(3)='file3', weight(3)=-1.d0 iflag=?? output_format=?? / where you must substitute ?? according to your needs and maybe add other keywords depending if you want 1D, 2D, ... plots HTH GS > presently i am working on Ni-Al-Ti system. I want to see the effect of > Ti in Ni-Al system. > I thought these plots could be of some help. > > -- > Regards > > K V Vamsi > Materials Engineering > Deformation mechanisms and modelling group > IISc, Bangalore - 560 012 > Mob : 09886818221 > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From mansourehp at gmail.com Wed May 6 09:26:08 2009 From: mansourehp at gmail.com (Mansoureh Pashangpour) Date: Wed, 6 May 2009 10:56:08 +0330 Subject: [Pw_forum] GGA functional Message-ID: Dear all I perform spin polarized GGA calculation for a system. I would like to know the GGA functional in this calculation and the order of density's partial in it. how can I obtain this informations? regards mansoureh -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090506/2755064f/attachment.htm From paulatto at sissa.it Wed May 6 10:04:58 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 06 May 2009 10:04:58 +0200 Subject: [Pw_forum] GGA functional In-Reply-To: References: Message-ID: In data 06 maggio 2009 alle ore 09:26:08, Mansoureh Pashangpour ha scritto: > Dear all > I perform spin polarized GGA calculation for a system. I would like to > know the GGA functional in this calculation Dear Mansoureh, the functional used by the code depends on the pseudopotentials. As they have to be consistent with each other, just open one of them with a text editor and look for it, it maybe something like: SLA PW PBE PBE PBE Exchange-Correlation functional the first four fields describe the model used for exchange, correlation, gradient correction for the exchange and for the correlation respectively; the rest is the common name for it, in this case "PBE Exchange-Correlation functional" There are other possible formats, though, but usually with a bit of imagination you can understand which functional is being used. Furthermore, if you've taken the pseudopotential file from the on-line library, the XC functional is specified in the file name, according to the convention explained in "Doc/nomefile.upf". > and the order of density's partial in it. If I understand correctly what you are asking for, it is zero (e.g. no derivative) for *non* gradient-corrected functionals (NOGX NOGC specified inside the file), and one for gradient-corrected ones. There are no meta-GGA pseudos, normal GGA ones are used instead. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From cyrille.barreteau at cea.fr Wed May 6 11:38:26 2009 From: cyrille.barreteau at cea.fr (Cyrille Barreteau) Date: Wed, 06 May 2009 11:38:26 +0200 Subject: [Pw_forum] dos In-Reply-To: <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> Message-ID: <4A015A92.7010408@cea.fr> Dear Stefano, Of course if you take the extreme case of the free electron gas, the local PDOS is obviously not more relevant than asking what it is s-like in Italy:-) (but maybe we could discover some interesting pertubation around Berlusconi brain:-) But in many cases the local PDOS can be very useful to analyze DFT results: -local magnetism -surface reactivity not to mention transport properties (STM images etc....) cyrille -- ================================================================== Cyrille Barreteau | phone : +33 (0)1 69 08 29 51 CEA Saclay | fax : +33 (0)1 69 08 84 46 DSM/IRAMIS/SPCSI | email cyrille.barreteau at cea.fr Batiment 462 | 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ http://iramis.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================================== Stefano Baroni wrote: > I am still rather skeptical. If you apply your reasoning to the free > electron gas, it seems to me that you would obtain rather weird (to > say the least) results. I may be wong, and finally the choice on how > to analyze your data is yours. SB > > On May 5, 2009, at 9:26 AM, Gabriele Sclauzero wrote: > >> >> Stefano Baroni wrote: >> >>> Jonas: I am afraid that what you want to obtain does not make much >> >>> sense. The angular character of any molecular (or Bloch) state depends >> >>> on the origin of your reference system. What is "s-like" with >>> respect to >> >>> one point has an infinite number of angular components with respect >>> to a >> >>> different point. Are you sure you want to add "what is s-like in Iowa" >> >>> with "what is s-like in Italy" ??? ;-) >> >>> Stefano >> >> >> I agree with what you say, Stefano, about taking the expectation >> value of the angular >> momentum in a periodic system, but summing the PDOS of selected atoms >> can lead to >> meaningful and sometimes very useful results, in some cases. >> I usually see the PDOS as an indicator of how much the Bloch states >> retain a kind of >> "atomic character" in the solid (so that in this case summing PDOS of >> equivalent atoms is >> the more meaningful thing, as Lex pointed out), but also as a way to >> identify the symmetry >> of states. >> If you place a molecule, a nanowire or maybe even a surface with slab >> geometry in a wise >> manner in your (super-)cell you can get many information from the >> PDOS (summed or not). >> Obviously, it does not make much sense to sum the s,p,d PDOS of all >> atoms in the system in >> general (but in some cases it may), so that's why you cannot find it >> directly in the >> output from projwfc. >> It's your task to select over which atoms you can do a sum (or an >> average) of the PDOS >> (and also placing the atoms in a smart way, first). I think you can >> write a script in >> awk/perl/whatever, or even a small fortran program, which reads PDOS >> from individual files >> associated to the selected atoms and sums the corresponding rows. >> >> Gabriele >> >>> >>> On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: >> >>> >>>> >>>> I just successfully calculated DOS and PDOS of a very large system. >>> >>>> One thing that confuses me is the presence of s, p components to PDOS >>> >>>> for every atom, but not summed (or averaged) s and p contribution of >>> >>>> all system. I would be happy to find in my output total DOS, PDOS s >>> >>>> character and PDOS p character. Is there a way to get that instead of >>> >>>> contribution per every atom? Or how do I proceed to get s and p PDOS >>> >>>> for all system by manually processing all of my 76 atoms PDOS >>>> outputs?.. >>> >>>> >>>> thanks >>> >>>> >>>> Jonas Baltrusaitis >>> >>>> University of Iowa >>> >>>> > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / > stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une morale > de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > >------------------------------------------------------------------------ > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From dat36 at cam.ac.uk Wed May 6 14:30:09 2009 From: dat36 at cam.ac.uk (David Tompsett) Date: Wed, 06 May 2009 13:30:09 +0100 Subject: [Pw_forum] Documentation on El-Ph coupling Message-ID: <4A0182D1.7010805@cam.ac.uk> Dear All, I am interested in calculating the electron-phonon coupling of a metallic system in QE. I would like to then apply the McMillan formula for the prediction of the superconducting Tc. In the basic tutorials list, /Hands-on Tutorial of Quantum Espresso ,/ I could find examples on the El-ph calculation for Aluminium at X. Is there any further documentation available to indicated on how to find the integrated lambda and wlog that enter the McMillan formula for Tc? Thank you, David. -- David A. Tompsett Quantum Matter Group Cavendish Laboratory J. J. Thomson Avenue Cambridge CB3 0HE U.K. Tel: +44 7907 566351 (mobile) Fax: +44 1223 768140 http://www-qm.phy.cam.ac.uk/ From jasius_1 at yahoo.com Wed May 6 15:01:00 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Wed, 6 May 2009 06:01:00 -0700 (PDT) Subject: [Pw_forum] submits jobs to node already in use In-Reply-To: <1241580589.10739.4.camel@zero> Message-ID: <848916.87796.qm@web54409.mail.yahoo.com> Axel, I mentioned PBS so qsub was implied. Machinefile is where it's supposed to be mpiexec -n 4 -machinefile $PBS_NODEFILE So I am not sure what's happening, if I submit say cpmd job it just goes to a different node, whereas pwscf to the same which already has a job running If I submit another pwscf with a simple script, not the one I adopted from Examples 2 and 9, it submits it to an independent node. It must be within my script, I guess Rocks forum was right after all. I'll have to look through it carefully Jonas --- On Tue, 5/5/09, Axel Kohlmeyer wrote: > From: Axel Kohlmeyer > Subject: Re: [Pw_forum] submits jobs to node already in use > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 5, 2009, 8:29 PM > On Tue, 2009-05-05 at 17:30 -0700, Jonas Baltrusaitis wrote: > > Funny story, if I submit any job with mpiexec -n 4 > -machinefile > > where is the machinefile?? > > > /../pw.x -rmpool 0 -nodes 1 -procs 4 via PBS to a 4 > core processor I > > this must be a specialty of how your machine is set up. if > you submit > to a PBS queue, you don't use mpiexec, but rather use > qsub. > mpiexec (or mpirun) is being called from within the job > script > that you submit to qsub. > > > can't see the exact number of cores running, e.g. > with qstat -n I see > > only compute-0-1/0, instead of usual > compute-0-1/0, compute-0-1/1, > > compute-0-1/2, compute-0-1/4. > > again. this has nothing at all to do with Q-E. there is no > code in > Q-E that knows anything about any batch system (and it > should not). > > > It gets even worse if I want to submit any other pwscf > job to my 16 > > node cluster. It get's submitted to exactly the > same node, e.g. I run > > 8 processes on a 4 core node and obviously no > performance > > > > I inquired at Rocks forum, they claim it must be > related with how Q-E > > submits jobs. Has anybody seen that before? > > that is wrong advice. > > cheers, > axel. > > > > > > Jonas Baltrusaitis > > University of Iowa > > > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of > Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, > PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: > 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a > better idiot. From akohlmey at cmm.chem.upenn.edu Wed May 6 15:44:15 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 06 May 2009 09:44:15 -0400 Subject: [Pw_forum] submits jobs to node already in use In-Reply-To: <848916.87796.qm@web54409.mail.yahoo.com> References: <848916.87796.qm@web54409.mail.yahoo.com> Message-ID: <1241617455.10739.102.camel@zero> On Wed, 2009-05-06 at 06:01 -0700, Jonas Baltrusaitis wrote: > Axel, I mentioned PBS so qsub was implied. jonas, please always be specific and never imply when posting messages to a mailing list. i have seen people doing extremely non-sensical things (in good faith), so it is impossible to tell up front, if something is implied or not. more importantly, in these issues details matter... a _lot_. if you leave them out, you restrict the accuracy of the answer you get. you _do_ want to get an accurate answer, do you? > Machinefile is where it's supposed to be mpiexec -n 4 -machinefile $PBS_NODEFILE > > So I am not sure what's happening, if I submit say cpmd job it just > goes to a different node, whereas pwscf to the same which already has > a job running i'd have to see the qsub script and know the qsub command line. but as a matter of principle. the node assignment is done by the resource manager and/or the MPI frontend command. _never_ by the application itself. i very much dislike the use of machine files to begin with. you should check whether the MPI library that you are using actually supports the TM job launch mechanism of torque/pbs. in that case you don't have to specify any machine file at all, _and_ have the added benefit that the resource manager will watch and kill all MPI tasks. two possible options are: you have a typo in your script. or you have multiple MPI implementation installed and you are launching an application compiled against one of them with a launch script of the other. > If I submit another pwscf with a simple script, not the one I adopted > from Examples 2 and 9, it submits it to an independent node. It must > be within my script, I guess Rocks forum was right after all. I'll > have to look through it carefully i disagree. the shell scripts that you write have nothing to do with the application itself. the job scripts shipped with Q-E are meant for interactive use. if you adapt them for batch use, you are at your own. particularly the -machinefile or equivalent flag is effectively something that works on good faith, but not on control. ask any sysadmin at a national center through which pains they go to confine parallel jobs to the nodes that have allocated. cheers, axel. > Jonas > > > --- On Tue, 5/5/09, Axel Kohlmeyer wrote: > > > From: Axel Kohlmeyer > > Subject: Re: [Pw_forum] submits jobs to node already in use > > To: jasius_1 at yahoo.com, "PWSCF Forum" > > Date: Tuesday, May 5, 2009, 8:29 PM > > On Tue, 2009-05-05 at 17:30 -0700, Jonas Baltrusaitis wrote: > > > Funny story, if I submit any job with mpiexec -n 4 > > -machinefile > > > > where is the machinefile?? > > > > > /../pw.x -rmpool 0 -nodes 1 -procs 4 via PBS to a 4 > > core processor I > > > > this must be a specialty of how your machine is set up. if > > you submit > > to a PBS queue, you don't use mpiexec, but rather use > > qsub. > > mpiexec (or mpirun) is being called from within the job > > script > > that you submit to qsub. > > > > > can't see the exact number of cores running, e.g. > > with qstat -n I see > > > only compute-0-1/0, instead of usual > > compute-0-1/0, compute-0-1/1, > > > compute-0-1/2, compute-0-1/4. > > > > again. this has nothing at all to do with Q-E. there is no > > code in > > Q-E that knows anything about any batch system (and it > > should not). > > > > > It gets even worse if I want to submit any other pwscf > > job to my 16 > > > node cluster. It get's submitted to exactly the > > same node, e.g. I run > > > 8 processes on a 4 core node and obviously no > > performance > > > > > > I inquired at Rocks forum, they claim it must be > > related with how Q-E > > > submits jobs. Has anybody seen that before? > > > > that is wrong advice. > > > > cheers, > > axel. > > > > > > > > > > Jonas Baltrusaitis > > > University of Iowa > > > > > > > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > ======================================================================= > > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > > http://www.cmm.upenn.edu > > Center for Molecular Modeling -- University of > > Pennsylvania > > Department of Chemistry, 231 S.34th Street, Philadelphia, > > PA 19104-6323 > > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: > > 1-215-898-5425 > > ======================================================================= > > If you make something idiot-proof, the universe creates a > > better idiot. > > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From paulatto at sissa.it Wed May 6 16:51:35 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 06 May 2009 16:51:35 +0200 Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <4A0182D1.7010805@cam.ac.uk> References: <4A0182D1.7010805@cam.ac.uk> Message-ID: In data 06 maggio 2009 alle ore 14:30:09, David Tompsett ha scritto: > Is there any > further documentation available to indicated on how to find the > integrated lambda and wlog that enter the McMillan formula for Tc? Dear David, example07 covers the electron-phono coupling calculations, I have no real practice with it, but I remember helping a friend doing such a calculation starting only from the original paper and that example... hence it should be possible. Feel free to ask if some point is not clear, regards. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From eyvaz_isaev at yahoo.com Wed May 6 17:23:18 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 6 May 2009 08:23:18 -0700 (PDT) Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <4A0182D1.7010805@cam.ac.uk> Message-ID: <909960.65607.qm@web65705.mail.ac4.yahoo.com> ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Wed, 5/6/09, David Tompsett wrote: > From: David Tompsett > Subject: [Pw_forum] Documentation on El-Ph coupling > To: pw_forum at pwscf.org > Date: Wednesday, May 6, 2009, 4:30 PM > Dear All, > I am interested in calculating the > electron-phonon coupling > of a metallic system in QE. I would like to then apply the > McMillan > formula for the prediction of the superconducting Tc. > > In the basic tutorials list, /Hands-on Tutorial of Quantum > Espresso > ,/ > I could find > examples on the El-ph calculation for Aluminium at X. Is > there any > further documentation available to indicated on how to find > the > integrated lambda and wlog that enter the McMillan formula > for Tc? > > Thank you, > David. > > -- > David A. Tompsett > Quantum Matter Group > Cavendish Laboratory > J. J. Thomson Avenue > Cambridge CB3 0HE > U.K. > Tel: +44 7907 566351 (mobile) > Fax: +44 1223 768140 > http://www-qm.phy.cam.ac.uk/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From dat36 at cam.ac.uk Wed May 6 18:12:57 2009 From: dat36 at cam.ac.uk (David Tompsett) Date: Wed, 06 May 2009 17:12:57 +0100 Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: References: <4A0182D1.7010805@cam.ac.uk> Message-ID: <4A01B709.3070700@cam.ac.uk> Dear Lorenzo, thank you for the advice. The example07 looks very helpful. Which is the original paper on the method that I should read? Thank you, David. Lorenzo Paulatto wrote: > In data 06 maggio 2009 alle ore 14:30:09, David Tompsett > ha scritto: > >> Is there any >> further documentation available to indicated on how to find the >> integrated lambda and wlog that enter the McMillan formula for Tc? >> > > Dear David, > example07 covers the electron-phono coupling calculations, I have no real > practice with it, but I remember helping a friend doing such a calculation > starting only from the original paper and that example... hence it should > be possible. > > Feel free to ask if some point is not clear, regards. > > -- David A. Tompsett Quantum Matter Group Cavendish Laboratory J. J. Thomson Avenue Cambridge CB3 0HE U.K. Tel: +44 7907 566351 (mobile) Fax: +44 1223 768140 http://www-qm.phy.cam.ac.uk/ From paulatto at sissa.it Wed May 6 18:40:08 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 06 May 2009 18:40:08 +0200 Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <4A01B709.3070700@cam.ac.uk> References: <4A0182D1.7010805@cam.ac.uk> <4A01B709.3070700@cam.ac.uk> Message-ID: In data 06 maggio 2009 alle ore 18:12:57, David Tompsett ha scritto: > Dear Lorenzo, thank you for the advice. The example07 looks very > helpful. Which is the original paper on the method that I should read? I was reading from this one: P.B. Allen, PRL 59, 1460 -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From baroni at sissa.it Wed May 6 18:59:32 2009 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 6 May 2009 18:59:32 +0200 Subject: [Pw_forum] vc-relax In-Reply-To: <295302.54036.qm@web33104.mail.mud.yahoo.com> References: <295302.54036.qm@web33104.mail.mud.yahoo.com> Message-ID: why don't you let it be? the calculated pressure is exceedingly small .. SB On May 6, 2009, at 8:23 AM, ali kazempoor wrote: > > Dear All > I want to study an O vacancy in 2*2*3 supercell for TiO2. I use from > vc-relax to optimize the lattice constant either with atomic > relaxation. But from one week ago, the pressure evolved around 0.00 > and dident go to zero .for example > (kbar) P= 0.07 > (kbar) P= -0.45 > (kbar) P= -0.23 > (kbar) P= -0.25 > (kbar) P= -0.34 > (kbar) P= -0.40 > (kbar) P= -0.36 > (kbar) P= -0.14 > (kbar) P= -0.57 > (kbar) P= -0.41 > (kbar) P= -0.45 > (kbar) P= -0.83 > (kbar) P= -0.45 > (kbar) P= -0.37 > (kbar) P= -1.58 > (kbar) P= -1.00 > (kbar) P= -0.71 > (kbar) P= -0.73 > (kbar) P= -0.61 > (kbar) P= -0.58 > (kbar) P= -0.39 > (kbar) P= -0.89 > (kbar) P= -0.58 > (kbar) P= -0.33 > (kbar) P= -0.37 > (kbar) P= 0.59 > (kbar) P= -0.42 > (kbar) P= -0.61 > (kbar) P= -1.22 > (kbar) P= -1.68 > (kbar) P= -0.23 > (kbar) P= -0.20 > (kbar) P= -0.17 > (kbar) P= -0.39 > (kbar) P= -0.29 > (kbar) P= -0.76 > (kbar) P= -0.30 > (kbar) P= -0.24 > what can I do for this problem? > Here is input file: > > > > &control > calculation = 'vc-relax' > restart_mode='restart', > prefix='rutile-71atom' > tstress = .true. > tprnfor = .true. > pseudo_dir = './', > outdir='/ptmp/javad/scratch/' > nstep=1000, > dt=100, > / > &system > ibrav= 6 , celldm(1) =17.36,celldm(3)=0.9663585, nat= 71, ntyp= > 2, > ecutwfc =35.0,ecutrho=400 > / > &electrons > diagonalization='' > mixing_mode = 'plain' > mixing_beta = 0.7 > conv_thr = 1.0d-4 > / > &IONS > / > &CELL > cell_dynamics = 'damp-w', > press=00.0 > / > ATOMIC_SPECIES > Ti 43.0 Ti.pz-sp-van.UPF > O 16.0 O.pz-van_ak.UPF > ATOMIC_POSITIONS crystal > Ti 0.00000000 0.00000000 0.00000000 > Ti 0.50000000 0.00000000 0.00000000 > Ti 0.00000000 0.50000000 0.00000000 > Ti 0.50000000 0.50000000 0.00000000 > Ti 0.00000000 0.00000000 0.33333333 > Ti 0.50000000 0.00000000 0.33333333 > Ti 0.00000000 0.50000000 0.33333333 > Ti 0.50000000 0.50000000 0.33333333 > Ti 0.00000000 0.00000000 0.66666667 > Ti 0.50000000 0.00000000 0.66666667 > Ti 0.00000000 0.50000000 0.66666667 > Ti 0.50000000 0.50000000 0.66666667 > Ti 0.25000000 0.25000000 0.16666667 > Ti 0.75000000 0.25000000 0.16666667 > Ti 0.25000000 0.75000000 0.16666667 > Ti 0.75000000 0.75000000 0.16666667 > Ti 0.25000000 0.25000000 0.50000000 > Ti 0.75000000 0.25000000 0.50000000 > Ti 0.25000000 0.75000000 0.50000000 > Ti 0.75000000 0.75000000 0.50000000 > Ti 0.25000000 0.25000000 0.83333333 > Ti 0.75000000 0.25000000 0.83333333 > Ti 0.25000000 0.75000000 0.83333333 > Ti 0.75000000 0.75000000 0.83333333 > O 0.65265000 0.15265000 0.00000000 > O 0.15265000 0.65265000 0.00000000 > O 0.65265000 0.65265000 0.00000000 > O 0.15265000 0.15265000 0.33333333 > O 0.65265000 0.15265000 0.33333333 > O 0.15265000 0.65265000 0.33333333 > O 0.65265000 0.65265000 0.33333333 > O 0.15265000 0.15265000 0.66666667 > O 0.65265000 0.15265000 0.66666667 > O 0.15265000 0.65265000 0.66666667 > O 0.65265000 0.65265000 0.66666667 > O 0.34735000 0.34735000 0.00000000 > O 0.84735000 0.34735000 0.00000000 > O 0.34735000 0.84735000 0.00000000 > O 0.84735000 0.84735000 0.00000000 > O 0.34735000 0.34735000 0.33333333 > O 0.84735000 0.34735000 0.33333333 > O 0.34735000 0.84735000 0.33333333 > O 0.84735000 0.84735000 0.33333333 > O 0.34735000 0.34735000 0.66666667 > O 0.84735000 0.34735000 0.66666667 > O 0.34735000 0.84735000 0.66666667 > O 0.84735000 0.84735000 0.66666667 > O 0.40265000 0.09735000 0.16666667 > O 0.90265000 0.09735000 0.16666667 > O 0.40265000 0.59735000 0.16666667 > O 0.90265000 0.59735000 0.16666667 > O 0.40265000 0.09735000 0.50000000 > O 0.90265000 0.09735000 0.50000000 > O 0.40265000 0.59735000 0.50000000 > O 0.90265000 0.59735000 0.50000000 > O 0.40265000 0.09735000 0.83333333 > O 0.90265000 0.09735000 0.83333333 > O 0.40265000 0.59735000 0.83333333 > O 0.90265000 0.59735000 0.83333333 > O 0.09735000 0.40265000 0.16666667 > O 0.59735000 0.40265000 0.16666667 > O 0.09735000 0.90265000 0.16666667 > O 0.59735000 0.90265000 0.16666667 > O 0.09735000 0.40265000 0.50000000 > O 0.59735000 0.40265000 0.50000000 > O 0.09735000 0.90265000 0.50000000 > O 0.59735000 0.90265000 0.50000000 > O 0.09735000 0.40265000 0.83333333 > O 0.59735000 0.40265000 0.83333333 > O 0.09735000 0.90265000 0.83333333 > O 0.59735000 0.90265000 0.83333333 > K_POINTS automatic > 3 3 2 1 1 1 > thanks alot > Ali Kazempour,Isfahan University of Technology > > > > > > > > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090506/eb703cc4/attachment-0001.htm From baroni at sissa.it Wed May 6 19:03:23 2009 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 6 May 2009 19:03:23 +0200 Subject: [Pw_forum] dos In-Reply-To: <4A015A92.7010408@cea.fr> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> Message-ID: Cyrille: I totally agree with you, of course. What I was questioning is the relevance of taking averages of pdos centered around different centers, not the concept of PDOS itself . Concerning Berlusconi's brain, I think that's perfectly OK. What is probably not is the brain of those who find his behavior appropriate. SB On May 6, 2009, at 11:38 AM, Cyrille Barreteau wrote: > Dear Stefano, > > Of course if you take the extreme case of the free electron gas, the > local PDOS is obviously not more relevant > than asking what it is s-like in Italy:-) (but maybe we could discover > some interesting pertubation around Berlusconi brain:-) > But in many cases the local PDOS can be very useful to analyze DFT > results: > -local magnetism > -surface reactivity > not to mention transport properties (STM images etc....) > > cyrille > > -- > ================================================================== > Cyrille Barreteau | phone : +33 (0)1 69 08 29 51 > CEA Saclay | fax : +33 (0)1 69 08 84 46 > DSM/IRAMIS/SPCSI | email cyrille.barreteau at cea.fr > Batiment 462 | > 91191 Gif sur Yvette Cedex FRANCE > ~~~~~~~~~~~~~~~~~~~~~~~~ > http://iramis.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html > ================================================================== > > > > > > Stefano Baroni wrote: > >> I am still rather skeptical. If you apply your reasoning to the free >> electron gas, it seems to me that you would obtain rather weird (to >> say the least) results. I may be wong, and finally the choice on how >> to analyze your data is yours. SB >> >> On May 5, 2009, at 9:26 AM, Gabriele Sclauzero wrote: >> >>> >>> Stefano Baroni wrote: >>> >>>> Jonas: I am afraid that what you want to obtain does not make much >>> >>>> sense. The angular character of any molecular (or Bloch) state >>>> depends >>> >>>> on the origin of your reference system. What is "s-like" with >>>> respect to >>> >>>> one point has an infinite number of angular components with respect >>>> to a >>> >>>> different point. Are you sure you want to add "what is s-like in >>>> Iowa" >>> >>>> with "what is s-like in Italy" ??? ;-) >>> >>>> Stefano >>> >>> >>> I agree with what you say, Stefano, about taking the expectation >>> value of the angular >>> momentum in a periodic system, but summing the PDOS of selected >>> atoms >>> can lead to >>> meaningful and sometimes very useful results, in some cases. >>> I usually see the PDOS as an indicator of how much the Bloch states >>> retain a kind of >>> "atomic character" in the solid (so that in this case summing PDOS >>> of >>> equivalent atoms is >>> the more meaningful thing, as Lex pointed out), but also as a way to >>> identify the symmetry >>> of states. >>> If you place a molecule, a nanowire or maybe even a surface with >>> slab >>> geometry in a wise >>> manner in your (super-)cell you can get many information from the >>> PDOS (summed or not). >>> Obviously, it does not make much sense to sum the s,p,d PDOS of all >>> atoms in the system in >>> general (but in some cases it may), so that's why you cannot find it >>> directly in the >>> output from projwfc. >>> It's your task to select over which atoms you can do a sum (or an >>> average) of the PDOS >>> (and also placing the atoms in a smart way, first). I think you can >>> write a script in >>> awk/perl/whatever, or even a small fortran program, which reads PDOS >>> from individual files >>> associated to the selected atoms and sums the corresponding rows. >>> >>> Gabriele >>> >>>> >>>> On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: >>> >>>> >>>>> >>>>> I just successfully calculated DOS and PDOS of a very large >>>>> system. >>>> >>>>> One thing that confuses me is the presence of s, p components to >>>>> PDOS >>>> >>>>> for every atom, but not summed (or averaged) s and p >>>>> contribution of >>>> >>>>> all system. I would be happy to find in my output total DOS, >>>>> PDOS s >>>> >>>>> character and PDOS p character. Is there a way to get that >>>>> instead of >>>> >>>>> contribution per every atom? Or how do I proceed to get s and p >>>>> PDOS >>>> >>>>> for all system by manually processing all of my 76 atoms PDOS >>>>> outputs?.. >>>> >>>>> >>>>> thanks >>>> >>>>> >>>>> Jonas Baltrusaitis >>>> >>>>> University of Iowa >>>> >>>>> >> --- >> Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - >> Trieste >> http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / >> stefanobaroni (skype) >> >> La morale est une logique de l'action comme la logique est une morale >> de la pens?e - Jean Piaget >> >> Please, if possible, don't send me MS Word or PowerPoint attachments >> Why? See: http://www.gnu.org/philosophy/no-word-attachments.html >> >> >> >> >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090506/c6df1302/attachment.htm From eyvaz_isaev at yahoo.com Wed May 6 19:08:01 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 6 May 2009 10:08:01 -0700 (PDT) Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <4A0182D1.7010805@cam.ac.uk> Message-ID: <998756.50141.qm@web65701.mail.ac4.yahoo.com> Hi, Sorry for empty mail, unexpectedly I clicked on Send button. Well, I think attached files are what you need. The first one (modified elphon.f90) writes special files, and the second one uses these files to find lambda and then T_c via the Allen-Dynes equation (\omega_log instead of the Debye temperature). In fact, recently I suggested to change original elphon.f90 and lambda.f90 files with those (at least, include changes), but I do not know whether this suggestion was accepted by developers. Attached elphon.f90 worked for QE 4.0.4, at least. Any questions, please contact me. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Wed, 5/6/09, David Tompsett wrote: > From: David Tompsett > Subject: [Pw_forum] Documentation on El-Ph coupling > To: pw_forum at pwscf.org > Date: Wednesday, May 6, 2009, 4:30 PM > Dear All, > I am interested in calculating the > electron-phonon coupling > of a metallic system in QE. I would like to then apply the > McMillan > formula for the prediction of the superconducting Tc. > > In the basic tutorials list, /Hands-on Tutorial of Quantum > Espresso > ,/ > I could find > examples on the El-ph calculation for Aluminium at X. Is > there any > further documentation available to indicated on how to find > the > integrated lambda and wlog that enter the McMillan formula > for Tc? > > Thank you, > David. > > -- > David A. Tompsett > Quantum Matter Group > Cavendish Laboratory > J. J. Thomson Avenue > Cambridge CB3 0HE > U.K. > Tel: +44 7907 566351 (mobile) > Fax: +44 1223 768140 > http://www-qm.phy.cam.ac.uk/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- A non-text attachment was scrubbed... Name: elphon.tar.gz Type: application/x-gzip Size: 12260 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20090506/da1ea1e9/attachment-0001.bin From cyrille.barreteau at cea.fr Wed May 6 19:52:26 2009 From: cyrille.barreteau at cea.fr (BARRETEAU Cyrille) Date: Wed, 6 May 2009 19:52:26 +0200 Subject: [Pw_forum] =?iso-8859-1?q?RE=A0=3A__dos?= References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> Message-ID: <18E4AC9C2FAC504E994A6759DE5ABE8D949A3C@DIODON.extra.cea.fr> Cyrille: I totally agree with you, of course. What I was questioning is the relevance of taking averages of pdos centered around different centers, not the concept of PDOS itself =================== OK I missed the point. I agree that averaging local pdos over different centers might not be so relevant... =================== Concerning Berlusconi's brain, I think that's perfectly OK. What is probably not is the brain of those who find his behavior appropriate. =========== I agree that the most scary is that Italian are voting for him, as the American voted for Bush! But maybe Italy will find it's Obama one day:-) For sure Europe will need an Obama do defend science! =========== ================================================================== Cyrille Barreteau | phone : +33 (0)1 69 08 29 51 CEA Saclay | fax : +33 (0)1 69 08 84 46 DSM/IRAMIS/SPCSI | email cyrille.barreteau at cea.fr Batiment 462 | 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ http://iramis.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================================== -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/ms-tnef Size: 3036 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20090506/0aeca49c/attachment.bin From timtro at gmail.com Wed May 6 23:33:32 2009 From: timtro at gmail.com (Tim Teatro) Date: Wed, 6 May 2009 17:33:32 -0400 Subject: [Pw_forum] How does the damped ion dynamics work? Message-ID: <560bcca0905061433n17f2b53ah3dbaaa8c9d2b613f@mail.gmail.com> Hi all, I have a rather naive question about the damped ion dynamics. Is it simply Verlet integration with friction-like coefficient? I'd like to understand this because if it as simple as I said above, I might like to try this as an alternative to Nose thermostat to cool down a system at a well defined rate. Thank you in advance. Best regards, Timothy A.V. Teatro Faculty of Science University of Ontario Institute of Technology CANADA -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090506/d1569c46/attachment.htm From ynwuyang at gmail.com Thu May 7 00:43:24 2009 From: ynwuyang at gmail.com (yuning wu) Date: Wed, 6 May 2009 18:43:24 -0400 Subject: [Pw_forum] Empirical van-der-waals correction Message-ID: <8f1062e50905061543n60810969yab8458c7ab70460d@mail.gmail.com> Dear all, I am interested in the empirical Van der Waals correction to DFT recently. Based on the information from previous discussion, the code of doing that (espresso/CPV/vanderwaals.f90) has already been developed. My question is, is this file included when compiling the package? If yes, how to use this function? If not, what should I do to include this file during compiling? Thanks. Best, Yuning -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090506/4a6fea11/attachment.htm From akohlmey at cmm.chem.upenn.edu Thu May 7 01:27:59 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 06 May 2009 19:27:59 -0400 Subject: [Pw_forum] Empirical van-der-waals correction In-Reply-To: <8f1062e50905061543n60810969yab8458c7ab70460d@mail.gmail.com> References: <8f1062e50905061543n60810969yab8458c7ab70460d@mail.gmail.com> Message-ID: <1241652479.2967.18.camel@zero> On Wed, 2009-05-06 at 18:43 -0400, yuning wu wrote: > Dear all, > > I am interested in the empirical Van der Waals correction to DFT > recently. Based on the information from previous discussion, the code > of doing that (espresso/CPV/vanderwaals.f90) has already been > developed. My question is, is this file included when compiling the > package? If yes, how to use this function? If not, what should I do to > include this file during compiling? Thanks. dear yuning, this code was written for the FPMD branch of the CP code in Q-E, quite a while back and has only coefficients for C and H parametrized for poly-ethane hard-coded, if i remember correctly. if you want to use this method in Q-E, you will first have to spend some time programming, which in turn means some reading of the code, which would be easier if you understood some italian, as most comments seems to be in that language. the good news is, that the code flow of the empirical vdw correction is fairly straightforward to understand and to implement, so it should still be not impossible to make this work in a more general fashion. cheers, axel. > > Best, > Yuning > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From nemeth at anl.gov Thu May 7 03:26:20 2009 From: nemeth at anl.gov (Karoly Nemeth) Date: Wed, 6 May 2009 20:26:20 -0500 (CDT) Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: Hi, I see that QE 4.0.5 has been released. I was aware of a warning in the 4.0 series related to pseudopotencial generation (using ld1.x in the 'atomic' directory). This was related to the UPF format writing, if I remember well. Are those problems of pseudopotencial generation fixed in the new, 4.0.5 version? Thanks: Karoly *********************************************************************** Karoly Nemeth, Ph.D. Argonne National Laboratory 9700 S. Cass Ave. Argonne, IL 60439. Advanced Photon Source X-Ray Science Division Building 401, Room B4198 Tel. 630-252-5813 Fax: 630-252-3222 *********************************************************************** From vitruss at gmail.com Thu May 7 12:19:07 2009 From: vitruss at gmail.com (Vit) Date: Thu, 07 May 2009 17:19:07 +0700 Subject: [Pw_forum] OpenMPI or not? Message-ID: <4A02B59B.6000800@googlemail.com> Hi, I've got a PC with Linux and 1 Phenom II x4 processor (4 core cpu). Do I need OpenMPI to run 1 job in 4 threads? or I just can complile QE with AMD acml_mp library and use the parallel calculations? Koroteev Victor, Nikolaev Institute of Inorganic Chemistry SB RAS From vitruss at gmail.com Thu May 7 12:29:13 2009 From: vitruss at gmail.com (Vit) Date: Thu, 7 May 2009 17:29:13 +0700 Subject: [Pw_forum] Compilation issue (phenom II and ACML) In-Reply-To: <7b6913e90905031619l507d8df7ta81a53d8ec2e819b@mail.gmail.com> References: <11f280750905030536r1b52bd83x1df4a5d35ee3731d@mail.gmail.com> <7b6913e90905031619l507d8df7ta81a53d8ec2e819b@mail.gmail.com> Message-ID: <11f280750905070329n32ff52ddrdbdabc9b557aa8d7@mail.gmail.com> I found the solution... ./configure LDFLAGS="-L/opt/acml4.2.0/gfortran64/lib/ -lacml" works (for MP LDFLAGS="-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml_mp" , but as the result I have The following libraries have been found: BLAS_LIBS= LAPACK_LIBS= -llapack FFT_LIBS= Please check if this is what you expect. As I see QE tryes to use lapack from the system, but not ACML and FFT is from internal (DFLAGS = -D__GFORTRAN -D__FFTW -D__USE_INTERNAL_FFTW). ACML can provede blas, lapack, fft libraryes. Why did it recognize only blas? Best wishes, Koroteev Victor, Nikolaev Institute of Inorganic Chemistry SB RAS. On Mon, May 4, 2009 at 6:19 AM, Axel Kohlmeyer wrote: > On Sun, May 3, 2009 at 8:36 AM, Vit wrote: >> Good day, everyone! >> I'm trying to use PWSCF on my own computer with Phenom II x4 with Arch >> Linux onboard. The User guide says that I can get huge benefit from >> using ACML from AMD. >> I've downloaded and installed it. Now I'm trying to compile QE with >> this library. If I use just "configure" it uses internal BLAS, LAPACK >> and FFTW libraries. If I try >> LDFLAGS="-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml" it says: > > hmmm... if you want to run with MPI, you should not use > the multi-threaded ACML but the serial one. see the comments > on using OMP_NUM_THREADS=1 with intel's MKL v10. > >> checking build system type... x86_64-unknown-linux-gnu >> checking architecture... amd64 >> checking for Fortran 77 compiler default output file name... >> configure: error: Fortran 77 compiler cannot create executables >> See `config.log' for more details. >> >> What's wrong? > [...] > >> configure:1695: checking for Fortran 77 compiler default output file >> name >> configure:1698: mpif77 ?-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml >> conftest.f ?>&5 >> /usr/bin/ld: cannot find -lacml >> collect2: ld returned 1 exit status > > looks like your ACML installation does not work. did you test > it on a simpler program before? > > cheers, > ? axel > >> configure:1701: $? = 1 >> configure: failed program was: >> | ? ? ? program main >> | >> | ? ? ? end >> configure:1740: error: Fortran 77 compiler cannot create executables >> See `config.log' for more details. > > > > -- > ======================================================================= > Axel Kohlmeyer ? akohlmey at cmm.chem.upenn.edu ? http://www.cmm.upenn.edu > ?Center for Molecular Modeling ? -- ? University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > tel: 1-215-898-1582, ?fax: 1-215-573-6233, ?office-tel: 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a better idiot. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From akohlmey at cmm.chem.upenn.edu Thu May 7 12:38:31 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 07 May 2009 06:38:31 -0400 Subject: [Pw_forum] Compilation issue (phenom II and ACML) In-Reply-To: <11f280750905070329n32ff52ddrdbdabc9b557aa8d7@mail.gmail.com> References: <11f280750905030536r1b52bd83x1df4a5d35ee3731d@mail.gmail.com> <7b6913e90905031619l507d8df7ta81a53d8ec2e819b@mail.gmail.com> <11f280750905070329n32ff52ddrdbdabc9b557aa8d7@mail.gmail.com> Message-ID: <1241692711.2967.22.camel@zero> On Thu, 2009-05-07 at 17:29 +0700, Vit wrote: > I found the solution... > ./configure LDFLAGS="-L/opt/acml4.2.0/gfortran64/lib/ -lacml" works > (for MP LDFLAGS="-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml_mp" , but > as the result I have victor, have you checked the ACML docs? most likely, you have to add -lpthread to link with -lacml_mp. last time i checked, the threading support in -lacml_mp was quite poor, so i would not bother trying to make it work. > The following libraries have been found: > BLAS_LIBS= > LAPACK_LIBS= -llapack > FFT_LIBS= > Please check if this is what you expect. > > As I see QE tryes to use lapack from the system, but not ACML and FFT > is from internal (DFLAGS = -D__GFORTRAN -D__FFTW > -D__USE_INTERNAL_FFTW). > ACML can provede blas, lapack, fft libraryes. Why did it recognize only blas? the configure script only uses what works with the embedded test scripts and what is supported by Q-E. ACML has _a_ FFT, but the API is different from what is supported. you'd have to program an interface. but don't worry about that, the FFTW library (even the internal one) is quite efficient and beats "vendor optimized" versions on a regular basis. cheers, axel. > Best wishes, > Koroteev Victor, > Nikolaev Institute of Inorganic Chemistry SB RAS. > > On Mon, May 4, 2009 at 6:19 AM, Axel Kohlmeyer > wrote: > > On Sun, May 3, 2009 at 8:36 AM, Vit wrote: > >> Good day, everyone! > >> I'm trying to use PWSCF on my own computer with Phenom II x4 with Arch > >> Linux onboard. The User guide says that I can get huge benefit from > >> using ACML from AMD. > >> I've downloaded and installed it. Now I'm trying to compile QE with > >> this library. If I use just "configure" it uses internal BLAS, LAPACK > >> and FFTW libraries. If I try > >> LDFLAGS="-L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml" it says: > > > > hmmm... if you want to run with MPI, you should not use > > the multi-threaded ACML but the serial one. see the comments > > on using OMP_NUM_THREADS=1 with intel's MKL v10. > > > >> checking build system type... x86_64-unknown-linux-gnu > >> checking architecture... amd64 > >> checking for Fortran 77 compiler default output file name... > >> configure: error: Fortran 77 compiler cannot create executables > >> See `config.log' for more details. > >> > >> What's wrong? > > [...] > > > >> configure:1695: checking for Fortran 77 compiler default output file > >> name > >> configure:1698: mpif77 -L/opt/acml4.2.0/gfortran64_mp/lib/ -lacml > >> conftest.f >&5 > >> /usr/bin/ld: cannot find -lacml > >> collect2: ld returned 1 exit status > > > > looks like your ACML installation does not work. did you test > > it on a simpler program before? > > > > cheers, > > axel > > > >> configure:1701: $? = 1 > >> configure: failed program was: > >> | program main > >> | > >> | end > >> configure:1740: error: Fortran 77 compiler cannot create executables > >> See `config.log' for more details. > > > > > > > > -- > > ======================================================================= > > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > > Center for Molecular Modeling -- University of Pennsylvania > > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > > ======================================================================= > > If you make something idiot-proof, the universe creates a better idiot. > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Thu May 7 12:42:48 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 07 May 2009 06:42:48 -0400 Subject: [Pw_forum] OpenMPI or not? In-Reply-To: <4A02B59B.6000800@googlemail.com> References: <4A02B59B.6000800@googlemail.com> Message-ID: <1241692968.2967.27.camel@zero> On Thu, 2009-05-07 at 17:19 +0700, Vit wrote: > Hi, > > I've got a PC with Linux and 1 Phenom II x4 processor (4 core cpu). Do I > need OpenMPI to run 1 job in 4 threads? or I just can complile QE with > AMD acml_mp library and use the parallel calculations? you can run with the multi-threaded BLAS, but that will only give you some limited speedup. the MPI parallelization is much more efficient in almost all, except a few corner cases. we had a discussion on that subject here a while ago. the tricky thing is to find out, which of the different levels of parallelism, that Q-E offers is the most efficient for your calculations. it used to be that the parallelization across k-points used to be the most efficient one, but that has a larger memory footprint and will thus have a lower cache efficiency, particularly on AMD cpus. we had a long discussion on that subject here not so long ago with some benchmark tests from several people. please check the mailing list archives and see, if there is anything useful in it for you. cheers, axel. > > Koroteev Victor, > Nikolaev Institute of Inorganic Chemistry SB RAS > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From paulatto at sissa.it Thu May 7 12:54:35 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 7 May 2009 12:54:35 +0200 (CEST) Subject: [Pw_forum] Compilation issue (phenom II and ACML) In-Reply-To: <11f280750905070329n32ff52ddrdbdabc9b557aa8d7@mail.gmail.com> References: <11f280750905030536r1b52bd83x1df4a5d35ee3731d@mail.gmail.com> <7b6913e90905031619l507d8df7ta81a53d8ec2e819b@mail.gmail.com> <11f280750905070329n32ff52ddrdbdabc9b557aa8d7@mail.gmail.com> Message-ID: <51854.78.12.168.42.1241693675.squirrel@webmail.sissa.it> On Thu, May 7, 2009 12:29, Vit wrote: > ACML can provede blas, lapack, fft libraryes. Why did it recognize only > blas? For some reason they are not recognised by configure, it happens quite often with acml. Nevertheless you can set make.sys manually in order to use all the features of acml. Here is how I do it: DFLAGS = -D__INTEL -D__MPI -D__PARA -D__ACML BLAS_LIBS = -L/path/acml/3.6.0/intel64/ifort64/lib -lacml -lacml_mv LAPACK_LIBS = FFT_LIBS = Note the -D__ACML in DFLAGS. It does not matter in which of the three *_LIB variables you put the acml inclusio, as they are all attached together in practice... With acml 4.x it does not seem to work for me. Performance improvement is not so impressive (but it may depend on system/hardware). regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From daniel.forrer at unipd.it Thu May 7 14:11:14 2009 From: daniel.forrer at unipd.it (Daniel Forrer) Date: Thu, 7 May 2009 14:11:14 +0200 Subject: [Pw_forum] Empirical van-der-waals correction In-Reply-To: <8f1062e50905061543n60810969yab8458c7ab70460d@mail.gmail.com> References: <8f1062e50905061543n60810969yab8458c7ab70460d@mail.gmail.com> Message-ID: Dear yuning, We have implemented the semi-empirical dispersion correction given by Grimme in the PWscf code (energy, forces and stress are available). Our modifications has been just sent to the developers and should be added soon to the CVS version. It works yet only with PWscf, but it should be not too hard to extend it to the CP. For further details please read http://www3.interscience.wiley.com/journal/121402787/abstract If you want to use our DFT-D implementation before it appears on CVS, please contact me to get the code. Best, daniel 2009/5/7 yuning wu > Dear all, > > I am interested in the empirical Van der Waals correction to DFT recently. > Based on the information from previous discussion, the code of doing that > (espresso/CPV/vanderwaals.f90) has already been developed. My question is, > is this file included when compiling the package? If yes, how to use this > function? If not, what should I do to include this file during compiling? > Thanks. > > Best, > Yuning > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ---------------------------------- Daniel Forrer, Ph.D Student Dipartimento di Scienze Chimiche Universit? degli Studi di Padova V. Marzolo 1, Padova mail daniel.forrer at unipd.it tel. +39 049 8275188 ---------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090507/dab251f5/attachment.htm From yccheng.nju at gmail.com Thu May 7 16:31:36 2009 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Thu, 7 May 2009 22:31:36 +0800 Subject: [Pw_forum] How to do vc-relax in CP code? Message-ID: Dear all, In the namelist of control, it seems that CP code can do a lot of calculations(scf, nscf,relax,vc-relax,vc-cp,...). I have tested the example18, which shows how to use cp.x to perform molecular dynamics simulation of SiO2. The first step is to get the ground state of the electronic system. I think there should be a previous step to get the equilibrium structure of SiO2. However I find there isn't any examples to do vc-relax using CP code. If I want to CP code to do vc-relax calculation, how to set corresponding parameters? In PW code, the vc-relax calculation is done in temperature=0 K condition which ignore the kinetic energy of ions. But in CP code, the vc-relax calculation complains that the temperature can not be 0 K. So I set the temperatures of ions and cell are 'not_controlled'. In the 'not_controlled' calculation, the cell of the system expands with the increasing steps. It is strange. How to do vc-relax in CP code? The following is the input file and some information from output file. input &control calculation='vc-relax', restart_mode='from_scratch', nstep=100, iprint=1, isave=1, dt=8.0, ndr=91, ndw=92, pseudo_dir='./', outdir='./', / &system ibrav=8, celldm(1)=9.28990, celldm(2)=1.73206, celldm(3)=1.09955, nat=18, ntyp=2, nbnd=48, nelec=96, nspin=1, ecutwfc=20.0, ecutrho=150.0, nr1b=16, nr2b=16, nr3b=16, qcutz=150., q2sigma=2.0, ecfixed=16.0, / &electrons electron_dynamics='verlet', ortho_max=100, emass=700., emass_cutoff=3., / &ions ion_dynamics='verlet', ion_radius(1)=1.0, ion_radius(2)=1.0, / &cell cell_dynamics='pr', press=0.D0 / ATOMIC_SPECIES O 16.00 O.pz-rrkjus.UPF Si 28.00 Si.vbc.UPF ATOMIC_POSITIONS O 3.18829368 14.83237039 1.22882961 O 7.83231469 6.78704039 1.22882961 O 2.07443467 5.99537992 4.73758250 O 6.72031366 14.04231898 4.73758250 O 3.96307134 11.26989826 7.87860582 O 8.60802134 3.22295920 7.87860582 O 3.96307134 4.81915267 9.14625133 O 8.60802134 12.86448267 9.14625133 O 3.18736469 1.25668055 5.58029607 O 7.83324368 9.30201055 5.58029607 O 2.07536366 10.09206195 2.07358613 O 6.71938467 2.04673195 2.07358613 Si 0.28891589 8.04533000 3.40456284 Si 4.93386589 0.00000000 3.40456284 Si 2.13389003 12.27717358 -0.04188031 Si 6.77884003 4.23184358 -0.04188031 Si 2.13389003 3.81348642 6.85202747 Si 6.77884003 11.85881642 6.85202747 some information form output file step 1 CELL_PARAMETERS 9.28990000 0.00000000 0.00000000 0.00000000 16.09066419 0.00000000 0.00000000 0.00000000 10.21470954 Total stress (GPa) 1582.97951906 -1.16929980 -21.68011687 -1.17872209 1618.59651136 11.14345397 -21.82861949 11.00777251 1647.98672617 step 50 CELL_PARAMETERS 16.77885645 -0.00946391 0.01423662 -0.01456595 21.79804238 -0.00448412 0.01470088 -0.00304592 17.53521234 Total stress (GPa) -11.78865891 0.58606252 1.97796694 0.58606252 95.04276174 -4.47537576 1.97796694 -4.47537576 7.30569481 step 100 CELL_PARAMETERS 26.25499658 -0.05592615 0.05092035 -0.07610163 31.84643730 0.00083894 0.05276954 0.00223134 27.45646153 Total stress (GPa) -21.80901098 0.11600975 -0.06986058 0.11600975 -1.12040472 0.25103238 -0.06986058 0.25103238 -19.41116556 -- Y. C. Cheng Department of Physics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090507/c21824f7/attachment-0001.htm From yccheng.nju at gmail.com Fri May 8 04:50:54 2009 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Fri, 8 May 2009 10:50:54 +0800 Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: Dear karoly, You can use ld1.x in 4.1CVS to generate UPF. Y. C. Cheng Department of Physics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/8d23be4d/attachment.htm From yaldachem at gmail.com Fri May 8 14:20:14 2009 From: yaldachem at gmail.com (yaldaa kh) Date: Fri, 8 May 2009 05:20:14 -0700 Subject: [Pw_forum] dos In-Reply-To: References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> Message-ID: <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> On 5/6/09, Stefano Baroni wrote: > > Cyrille: I totally agree with you, of course. What I was questioning is the > relevance of taking averages of pdos centered around different centers, not > the concept of PDOS itself . Concerning Berlusconi's brain, I think that's > perfectly OK. What is probably not is the brain of those who find his > behavior appropriate. > > > SB > > On May 6, 2009, at 11:38 AM, Cyrille Barreteau wrote: > > Dear Stefano, > > Of course if you take the extreme case of the free electron gas, the > local PDOS is obviously not more relevant > than asking what it is s-like in Italy:-) (but maybe we could discover > some interesting pertubation around Berlusconi brain:-) > But in many cases the local PDOS can be very useful to analyze DFT > results: > -local magnetism > -surface reactivity > not to mention transport properties (STM images etc....) > > cyrille > > -- > ================================================================== > Cyrille Barreteau | phone : +33 (0)1 69 08 29 51 > CEA Saclay | fax : +33 (0)1 69 08 84 46 > DSM/IRAMIS/SPCSI | email cyrille.barreteau at cea.fr > Batiment 462 | > 91191 Gif sur Yvette Cedex FRANCE > ~~~~~~~~~~~~~~~~~~~~~~~~ > http://iramis.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html > ================================================================== > > > > > > Stefano Baroni wrote: > > I am still rather skeptical. If you apply your reasoning to the free > > electron gas, it seems to me that you would obtain rather weird (to > > say the least) results. I may be wong, and finally the choice on how > > to analyze your data is yours. SB > > > On May 5, 2009, at 9:26 AM, Gabriele Sclauzero wrote: > > > > Stefano Baroni wrote: > > > Jonas: I am afraid that what you want to obtain does not make much > > > sense. The angular character of any molecular (or Bloch) state depends > > > on the origin of your reference system. What is "s-like" with > > respect to > > > one point has an infinite number of angular components with respect > > to a > > > different point. Are you sure you want to add "what is s-like in Iowa" > > > with "what is s-like in Italy" ??? ;-) > > > Stefano > > > > I agree with what you say, Stefano, about taking the expectation > > value of the angular > > momentum in a periodic system, but summing the PDOS of selected atoms > > can lead to > > meaningful and sometimes very useful results, in some cases. > > I usually see the PDOS as an indicator of how much the Bloch states > > retain a kind of > > "atomic character" in the solid (so that in this case summing PDOS of > > equivalent atoms is > > the more meaningful thing, as Lex pointed out), but also as a way to > > identify the symmetry > > of states. > > If you place a molecule, a nanowire or maybe even a surface with slab > > geometry in a wise > > manner in your (super-)cell you can get many information from the > > PDOS (summed or not). > > Obviously, it does not make much sense to sum the s,p,d PDOS of all > > atoms in the system in > > general (but in some cases it may), so that's why you cannot find it > > directly in the > > output from projwfc. > > It's your task to select over which atoms you can do a sum (or an > > average) of the PDOS > > (and also placing the atoms in a smart way, first). I think you can > > write a script in > > awk/perl/whatever, or even a small fortran program, which reads PDOS > > from individual files > > associated to the selected atoms and sums the corresponding rows. > > > Gabriele > > > > On May 4, 2009, at 5:51 PM, Jonas Baltrusaitis wrote: > > > > > I just successfully calculated DOS and PDOS of a very large system. > > > One thing that confuses me is the presence of s, p components to PDOS > > > for every atom, but not summed (or averaged) s and p contribution of > > > all system. I would be happy to find in my output total DOS, PDOS s > > > character and PDOS p character. Is there a way to get that instead of > > > contribution per every atom? Or how do I proceed to get s and p PDOS > > > for all system by manually processing all of my 76 atoms PDOS > > outputs?.. > > > > thanks > > > > Jonas Baltrusaitis > > > University of Iowa > > > > --- > > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > > http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / > > stefanobaroni (skype) > > > La morale est une logique de l'action comme la logique est une morale > > de la pens?e - Jean Piaget > > > Please, if possible, don't send me MS Word or PowerPoint attachments > > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > ------------------------------------------------------------------------ > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une morale de la > pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > dear all after such considerable discussion my problem to project dos to molecular (not atomic) orbital (specially antibonding ) has not solved. Please anyone explain to me. thanks yalda -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/d58bcfae/attachment.htm From paulatto at sissa.it Fri May 8 14:28:43 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 08 May 2009 14:28:43 +0200 Subject: [Pw_forum] dos In-Reply-To: <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> Message-ID: > dear all > after such considerable discussion my problem to project dos to molecular > (not atomic) orbital (specially antibonding ) has not solved. Please > anyone explain to me. Ok, let's review your original questions: > How can I project dos to molecular orbital? You cannot, there are no molecular orbitals defined in Quantum-Espresso. Implementing it is probably far from trivial, as molecular orbitals are not easy to define in the plane-wave formalism. > what is the importance of nbnd in calculations for dos? Only the bands that have been computed can be included in the DOS. If you want the DOS to be accurate well above the Fermi enegy, take care to include enough bands either directly in the scf calculation or doing an nscf calculation with more bands after the scf has completed. > and for plottingdos or pdos, if the fermi energy is assumed to be zero > automatically by program? No, it is not. You can get the Fermi energy (in eV) directly from the output of the scf calculation, if you are not using smearing you will get HOMO and LUMO energies instead. > thanks you're welcome regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From akohlmey at cmm.chem.upenn.edu Fri May 8 14:51:21 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 08 May 2009 08:51:21 -0400 Subject: [Pw_forum] dos In-Reply-To: References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> Message-ID: <1241787081.3356.8.camel@zero> On Fri, 2009-05-08 at 14:28 +0200, Lorenzo Paulatto wrote: > > How can I project dos to molecular orbital? > > You cannot, there are no molecular orbitals defined in Quantum-Espresso. > Implementing it is probably far from trivial, as molecular orbitals are > not easy to define in the plane-wave formalism. one comment. in a strict sense, there are no molecular orbitals in a periodic calculation. they become bands. however, people have tried to find an approximation that would work reasonably well for molecular crystals (you'd assume that the molecules don't interact, which they do). for this you first run a localization procedure and compute the wannier centers and orbitals, then you determine which atoms belong to a molecule and then distribute the wannier centers and orbitals to those molecules and diagonalize the resulting submatrices again. now in principle you could use those orbitals to project onto for a pDOS calculation. needless to say, that this all would require some programming to implement it in Q-E. cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From baroni at sissa.it Fri May 8 17:22:23 2009 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 8 May 2009 17:22:23 +0200 Subject: [Pw_forum] dos In-Reply-To: <1241787081.3356.8.camel@zero> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> <1241787081.3356.8.camel@zero> Message-ID: <7632F40A-DA5A-4C66-AA80-42B26F3E3DAB@sissa.it> I think the question was slightly different. It would make sense to project the DOS onto any localized orbital, and this irrespective of the basis set employed (PW's, Gaussians, you name it ...) and of the nature of the system (finite or extended), just in the same way as the pdos is defined with respect to an atomic orbital. Projecting the dos onto a molecular orbital is not implemenetd in QE, as far as I know. A quick-and-dirty way of achieving your goal (which would require some work, elas!) is the following: definition of PDOS(\omega) =\sum_n |<\psi_n|\phi_0>^2 \delta(omega- E_n), with a notation which should be obvious, A simple way to obtain \phi_0 in a form that would make it (relatively) easy to calculate the scalar products and the ensuing sum is to make a SCF calculation of the isolated molecule using a supercell of exactly the same geometry as used for the full system, using the same kinetic-energy cutoff, etc, so that the number of PWs, as well as the ordering of the G vectors would be the same. The aclculation of scalar products should be trivial, and the delta functions could be replaced by Lorentians or Gaussians of finite width. If you do this, maybe the resulting small program that you produce could be of some interests to other users in this mailing list, and you may want to share it with this community. Keep us informed of any progress (or lack thereof) Stefano On May 8, 2009, at 2:51 PM, Axel Kohlmeyer wrote: > On Fri, 2009-05-08 at 14:28 +0200, Lorenzo Paulatto wrote: > >>> How can I project dos to molecular orbital? >> >> You cannot, there are no molecular orbitals defined in Quantum- >> Espresso. >> Implementing it is probably far from trivial, as molecular orbitals >> are >> not easy to define in the plane-wave formalism. > > one comment. > > in a strict sense, there are no molecular orbitals > in a periodic calculation. they become bands. > > however, people have tried to find an approximation that would > work reasonably well for molecular crystals (you'd assume that > the molecules don't interact, which they do). for this you > first run a localization procedure and compute the wannier centers > and orbitals, then you determine which atoms belong to a molecule > and then distribute the wannier centers and orbitals to those > molecules and diagonalize the resulting submatrices again. > > now in principle you could use those orbitals to project onto > for a pDOS calculation. > > needless to say, that this all would require some programming > to implement it in Q-E. > > cheers, > axel. > > -- > = > ====================================================================== > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http:// > www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA > 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > = > ====================================================================== > If you make something idiot-proof, the universe creates a better > idiot. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/c8c1edc8/attachment.htm From baroni at sissa.it Fri May 8 17:28:05 2009 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 8 May 2009 17:28:05 +0200 Subject: [Pw_forum] dos In-Reply-To: <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> Message-ID: oh, by the way: as you probably know, people in this forum appreciate properly signed posts, including the full affiliation of the sender. Also, it would be a very good idea to write your post before and not after the sequel of attachments of previous posts. Stefano On May 8, 2009, at 2:20 PM, yaldaa kh wrote: > > ... > > dear all > after such considerable discussion my problem to project dos to > molecular (not atomic) orbital (specially antibonding ) has not > solved. Please anyone explain to me. > thanks > yalda > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/dde4f36b/attachment.htm From ceresoli at MIT.EDU Fri May 8 17:38:40 2009 From: ceresoli at MIT.EDU (Davide Ceresoli) Date: Fri, 08 May 2009 11:38:40 -0400 Subject: [Pw_forum] dos In-Reply-To: <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> Message-ID: <4A045200.1090404@mit.edu> yaldaa kh wrote: > > dear all > after such considerable discussion my problem to project dos to > molecular (not atomic) orbital (specially antibonding ) has not solved. > Please anyone explain to me. > thanks > yalda > > Dear Yalda, you can have a look at this paper: Vibrational recognition of adsorption sites for CO on platinum-and platinum-ruthenium surfaces Dabo I., Wieckowski A., Marzari N. Journal of the American Chemical Society 129, 11045-11052 (2007) DOI: http://dx.doi.org/10.1021/ja067944u HTH. Davide -- +----------------------------------------------------------+ Davide Ceresoli DMSE 13-4084 Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge, MA 02139-4307 Phone: (617) 253-6026 Mobile: +39-347-1001570 Skype: dceresoli +----------------------------------------------------------+ From jasius_1 at yahoo.com Fri May 8 17:46:51 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Fri, 8 May 2009 08:46:51 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <4A045200.1090404@mit.edu> Message-ID: <918309.11674.qm@web54401.mail.yahoo.com> Great paper. This is what I was inquiring earlier about summing up PDOS for specific elements (or orbitals in specific element, for that matter). So now I am confused: all previous posts convinced me that summing up PDOS for certain atoms is not the way to go, and in this paper that's exactly what has been done. Or am I missing something? Jonas Baltrusaitis university of Iowa --- On Fri, 5/8/09, Davide Ceresoli wrote: > From: Davide Ceresoli > Subject: Re: [Pw_forum] dos > To: "PWSCF Forum" > Date: Friday, May 8, 2009, 8:38 AM > yaldaa kh wrote: > > > > dear all > > after such considerable discussion my problem to > project dos to > > molecular (not atomic) orbital (specially antibonding > ) has not solved. > > Please anyone explain to me. > > thanks > > yalda > > > > > Dear Yalda, > you can have a look at this paper: > > Vibrational recognition of adsorption sites for CO on > platinum-and > platinum-ruthenium surfaces > Dabo I., Wieckowski A., Marzari N. > Journal of the American Chemical Society 129, 11045-11052 > (2007) > > DOI: http://dx.doi.org/10.1021/ja067944u > > HTH. > > Davide > > > -- > +----------------------------------------------------------+ > Davide Ceresoli > DMSE 13-4084 > Massachusetts Institute of Technology > 77 Massachusetts Avenue > Cambridge, MA 02139-4307 > Phone: (617) 253-6026 > Mobile: +39-347-1001570 > Skype: dceresoli > +----------------------------------------------------------+ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From akohlmey at cmm.chem.upenn.edu Fri May 8 17:55:54 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 08 May 2009 11:55:54 -0400 Subject: [Pw_forum] dos In-Reply-To: <7632F40A-DA5A-4C66-AA80-42B26F3E3DAB@sissa.it> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> <1241787081.3356.8.camel@zero> <7632F40A-DA5A-4C66-AA80-42B26F3E3DAB@sissa.it> Message-ID: <1241798154.2961.78.camel@zero> On Fri, 2009-05-08 at 17:22 +0200, Stefano Baroni wrote: > I think the question was slightly different. It would make sense to > project the DOS onto any localized orbital, and this irrespective of > the basis set employed (PW's, Gaussians, you name it ...) and of the > nature of the system (finite or extended), just in the same way as the > pdos is defined with respect to an atomic orbital. Projecting the dos > onto a molecular orbital is not implemenetd in QE, as far as I know. A > quick-and-dirty way of achieving your goal (which would require some > work, elas!) is the following: dear stefano, i agree with this assessment. i don't see a principal technical problem to get some _approximation_ of a molecular pDOS, but one would have to be very careful in interpreting the results. particularly the notion of anti-bonding MOs worries me. those tend to have restricted meaning in pseudopotential DFT calculations to begin with. now generating them from an approximation and then doing a projection on top of it, would make me worry about it being a very complicated way to generate quasi-random numbers. in any case one has to be well aware of the fact, that the molecular wavefunction that the DOS are projected on is to a significant degree arbitrarily chosen and it is based on an assumption of non-interacting molecules. that being said, i can certainly imagine situations where this kind of procedure would make sense. cheers, axel. -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Fri May 8 17:58:51 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 08 May 2009 11:58:51 -0400 Subject: [Pw_forum] dos In-Reply-To: <918309.11674.qm@web54401.mail.yahoo.com> References: <918309.11674.qm@web54401.mail.yahoo.com> Message-ID: <1241798331.2961.85.camel@zero> On Fri, 2009-05-08 at 08:46 -0700, Jonas Baltrusaitis wrote: > Great paper. This is what I was inquiring earlier about summing up PDOS > for specific elements (or orbitals in specific element, for that > matter). So now I am confused: all previous posts convinced me that > summing up PDOS for certain atoms is not the way to go, and in this > paper that's exactly what has been done. my interpretation of the discussion was, that you can sum up pDOS of chemically equivalent states (e.g. in a supercell), but not just arbitrarily anything. HTH, axel. > Or am I missing something? > > Jonas Baltrusaitis > university of Iowa > > > --- On Fri, 5/8/09, Davide Ceresoli wrote: > > > From: Davide Ceresoli > > Subject: Re: [Pw_forum] dos > > To: "PWSCF Forum" > > Date: Friday, May 8, 2009, 8:38 AM > > yaldaa kh wrote: > > > > > > dear all > > > after such considerable discussion my problem to > > project dos to > > > molecular (not atomic) orbital (specially antibonding > > ) has not solved. > > > Please anyone explain to me. > > > thanks > > > yalda > > > > > > > > Dear Yalda, > > you can have a look at this paper: > > > > Vibrational recognition of adsorption sites for CO on > > platinum-and > > platinum-ruthenium surfaces > > Dabo I., Wieckowski A., Marzari N. > > Journal of the American Chemical Society 129, 11045-11052 > > (2007) > > > > DOI: http://dx.doi.org/10.1021/ja067944u > > > > HTH. > > > > Davide > > > > > > -- > > +----------------------------------------------------------+ > > Davide Ceresoli > > DMSE 13-4084 > > Massachusetts Institute of Technology > > 77 Massachusetts Avenue > > Cambridge, MA 02139-4307 > > Phone: (617) 253-6026 > > Mobile: +39-347-1001570 > > Skype: dceresoli > > +----------------------------------------------------------+ > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From marzari at MIT.EDU Fri May 8 18:01:41 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Fri, 08 May 2009 12:01:41 -0400 Subject: [Pw_forum] dos In-Reply-To: <1241798154.2961.78.camel@zero> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> <1241787081.3356.8.camel@zero> <7632F40A-DA5A-4C66-AA80-42B26F3E3DAB@sissa.it> <1241798154.2961.78.camel@zero> Message-ID: <4A045765.3080505@mit.edu> Axel Kohlmeyer wrote: > On Fri, 2009-05-08 at 17:22 +0200, Stefano Baroni wrote: >> I think the question was slightly different. It would make sense to >> project the DOS onto any localized orbital, and this irrespective of >> the basis set employed (PW's, Gaussians, you name it ...) and of the >> nature of the system (finite or extended), just in the same way as the >> pdos is defined with respect to an atomic orbital. Projecting the dos >> onto a molecular orbital is not implemenetd in QE, as far as I know. Dear All, apologies about not having followed the pdos discussion - for the dos one, though, projection on maxloc Wannier functions was first discussed here (PL Silvestrelli et al, Solid State Communication, 1998) http://dx.doi.org/10.1016/S0038-1098(98)00175-6 It would be really easy to construct these data from the wannier code; we'll put this on the to do list. nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From jasius_1 at yahoo.com Fri May 8 18:06:11 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Fri, 8 May 2009 09:06:11 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <1241798331.2961.85.camel@zero> Message-ID: <701722.11609.qm@web54404.mail.yahoo.com> arbitrarily anything is meaningless. I was asking exactly about summing equivalent atom orbitals, say C2s or O2p. Now the paper below talks about a uniform surface. I am working on molecular crystals, where pretty much every atom will have different environment, repeated in 3-D. Would summing those PDOS for elements be meaningful then? Jonas --- On Fri, 5/8/09, Axel Kohlmeyer wrote: > From: Axel Kohlmeyer > Subject: Re: [Pw_forum] dos > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Friday, May 8, 2009, 8:58 AM > On Fri, 2009-05-08 at 08:46 -0700, Jonas Baltrusaitis wrote: > > Great paper. This is what I was inquiring earlier > about summing up PDOS > > for specific elements (or orbitals in specific > element, for that > > matter). So now I am confused: all previous posts > convinced me that > > summing up PDOS for certain atoms is not the way to > go, and in this > > paper that's exactly what has been done. > > my interpretation of the discussion was, that you can sum > up > pDOS of chemically equivalent states (e.g. in a supercell), > > but not just arbitrarily anything. > > HTH, > axel. > > > Or am I missing something? > > > > Jonas Baltrusaitis > > university of Iowa > > > > > > --- On Fri, 5/8/09, Davide Ceresoli > wrote: > > > > > From: Davide Ceresoli > > > Subject: Re: [Pw_forum] dos > > > To: "PWSCF Forum" > > > > Date: Friday, May 8, 2009, 8:38 AM > > > yaldaa kh wrote: > > > > > > > > dear all > > > > after such considerable discussion my > problem to > > > project dos to > > > > molecular (not atomic) orbital (specially > antibonding > > > ) has not solved. > > > > Please anyone explain to me. > > > > thanks > > > > yalda > > > > > > > > > > > Dear Yalda, > > > you can have a look at this paper: > > > > > > Vibrational recognition of adsorption sites for > CO on > > > platinum-and > > > platinum-ruthenium surfaces > > > Dabo I., Wieckowski A., Marzari N. > > > Journal of the American Chemical Society 129, > 11045-11052 > > > (2007) > > > > > > DOI: http://dx.doi.org/10.1021/ja067944u > > > > > > HTH. > > > > > > Davide > > > > > > > > > -- > > > > +----------------------------------------------------------+ > > > Davide Ceresoli > > > DMSE 13-4084 > > > Massachusetts Institute of Technology > > > 77 Massachusetts Avenue > > > Cambridge, MA 02139-4307 > > > Phone: (617) 253-6026 > > > Mobile: +39-347-1001570 > > > Skype: dceresoli > > > > +----------------------------------------------------------+ > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of > Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, > PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: > 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a > better idiot. From akohlmey at cmm.chem.upenn.edu Fri May 8 18:18:59 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 08 May 2009 12:18:59 -0400 Subject: [Pw_forum] dos In-Reply-To: <701722.11609.qm@web54404.mail.yahoo.com> References: <701722.11609.qm@web54404.mail.yahoo.com> Message-ID: <1241799539.2961.130.camel@zero> On Fri, 2009-05-08 at 09:06 -0700, Jonas Baltrusaitis wrote: > arbitrarily anything is meaningless. I was asking exactly about summing > equivalent atom orbitals, say C2s or O2p. but those two C 2s orbitals have to be in an _equivalent environment_, too. > Now the paper below talks about a uniform surface. I am working on > molecular crystals, where pretty much every atom will have different > environment, repeated in 3-D. Would summing those PDOS for elements be > meaningful then? not over elements per se. consider an ethanol molecule. the two carbons will have very different environments. axel. > Jonas > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From jasius_1 at yahoo.com Fri May 8 18:40:42 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Fri, 8 May 2009 09:40:42 -0700 (PDT) Subject: [Pw_forum] dos In-Reply-To: <1241799539.2961.130.camel@zero> Message-ID: <705538.55739.qm@web54401.mail.yahoo.com> affirmative. we are on the same track. this is so cool, thanks whomever showed this paper. I'll break down my molecular crystals into several components, such as "pyridine ring C2s", "hydroxyl O2s" etc. that will give wealth of information I've been meaning to ask but due to my inherent (_cough_) shyness didn't. In calculated PDS files, what's a ldos(E) and how it differs from pdos(E)? Furthermore, for p orbitals I get ldos and three pdos. I honestly went through DOS manuals but couldn't figure that out. At a risk of being referred to older post in the forum, I would ask clarification on what those mean thanks JOnas --- On Fri, 5/8/09, Axel Kohlmeyer wrote: > From: Axel Kohlmeyer > Subject: Re: [Pw_forum] dos > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Friday, May 8, 2009, 9:18 AM > On Fri, 2009-05-08 at 09:06 -0700, Jonas Baltrusaitis wrote: > > arbitrarily anything is meaningless. I was asking > exactly about summing > > equivalent atom orbitals, say C2s or O2p. > > but those two C 2s orbitals have to be in an > _equivalent environment_, too. > > > Now the paper below talks about a uniform surface. I > am working on > > molecular crystals, where pretty much every atom will > have different > > environment, repeated in 3-D. Would summing those > PDOS for elements be > > meaningful then? > > not over elements per se. consider an ethanol molecule. the > two > carbons will have very different environments. > > axel. > > > > Jonas > > > > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of > Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, > PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: > 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a > better idiot. From nemeth at anl.gov Fri May 8 19:09:53 2009 From: nemeth at anl.gov (Karoly Nemeth) Date: Fri, 8 May 2009 12:09:53 -0500 (CDT) Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: Thanks Y. C. Cheng! However, I do not see any listing of the 4.1 CVS version from the PWSCF download page or elsewhere. How can I get it? Thanks: Karoly > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 8 May 2009 10:50:54 +0800 > From: ??? > Subject: Re: [Pw_forum] QE 4.0.5 pseudopotencials > To: PWSCF Forum > Message-ID: > > Content-Type: text/plain; charset="utf-8" > > Dear karoly, > You can use ld1.x in 4.1CVS to generate UPF. > > Y. C. Cheng > Department of Physics > Nanjing University > Nanjing 210093 > P. R. China > Tel: 86-25-83592907 > Email: yccheng.nju at gmail.com > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/8d23be4d/attachment.html > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 23, Issue 19 > **************************************** > From cristian.degliesposti at unibo.it Fri May 8 19:25:46 2009 From: cristian.degliesposti at unibo.it (Cristian Degli Esposti Boschi) Date: Fri, 08 May 2009 19:25:46 +0200 Subject: [Pw_forum] Format of dynamical matrices Message-ID: <4A046B1A.5050809@unibo.it> Dear all, I have a couple of questions regarding the format of dynamical matrices generated by ph.x. (1) After the line "Dynamical Matrix in cartesian axes" and the q-value (0 0 0 in my cases) I see a block of 3 rows and 6 columns for each possible choice of say (n,m) where n and m label the atoms. What do the entries of the rows and columns represent? It seems that the second element of every entry is always 0... (2) After the line "Diagonalizing the dynamical matrix" and the q-value, enclosed in two lines of ***** I see the dynamical matrix expressed as eigenvalues/eigenvectors. Are the components of the latter expressed (for every atom) in real space? Again what are the zeroes after each component? I attach the output for fcc Si from example02. Thanks for your time, Cristian Dynamical matrix file 1 2 2 10.2000000 0.0000000 0.0000000 0.0000000 0.0000000 0.0000000 1 'Si ' 25593.354256961386 1 1 0.0000000 0.0000000 0.0000000 2 1 0.2500000 0.2500000 0.2500000 Dynamical Matrix in cartesian axes q = ( 0.000000000 0.000000000 0.000000000 ) 1 1 0.27657353 0.00000000 -0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.27657353 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.27657353 0.00000000 1 2 -0.27655447 0.00000000 -0.00000000 0.00000000 -0.00000000 0.00000000 -0.00000000 0.00000000 -0.27655447 0.00000000 0.00000000 0.00000000 -0.00000000 0.00000000 0.00000000 0.00000000 -0.27655447 0.00000000 2 1 -0.27655447 0.00000000 -0.00000000 0.00000000 -0.00000000 0.00000000 -0.00000000 0.00000000 -0.27655447 0.00000000 0.00000000 0.00000000 -0.00000000 0.00000000 0.00000000 0.00000000 -0.27655447 0.00000000 2 2 0.27657353 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 -0.00000000 0.00000000 0.27657353 0.00000000 0.00000000 0.00000000 -0.00000000 0.00000000 -0.00000000 0.00000000 0.27657353 0.00000000 Dielectric Tensor: 13.806429681482 0.000000000000 -0.000000000000 0.000000000000 13.806429681482 -0.000000000000 -0.000000000000 -0.000000000000 13.806429681482 Effective Charges E-U: Z_{alpha}{s,beta} atom # 1 -0.075684611458 0.000000000000 -0.000000000000 0.000000000000 -0.075684611458 0.000000000000 -0.000000000000 0.000000000000 -0.075684611458 atom # 2 -0.075684611458 -0.000000000000 -0.000000000000 -0.000000000000 -0.075684611458 0.000000000000 -0.000000000000 0.000000000000 -0.075684611458 Diagonalizing the dynamical matrix q = ( 0.000000000 0.000000000 0.000000000 ) ************************************************************************** omega( 1) = 0.089782 [THz] = 2.994825 [cm-1] ( -0.418364 0.000000 0.344966 0.000000 0.453839 0.000000 ) ( -0.418364 0.000000 0.344966 0.000000 0.453839 0.000000 ) omega( 2) = 0.089782 [THz] = 2.994825 [cm-1] ( 0.075065 0.000000 0.591377 0.000000 -0.380313 0.000000 ) ( 0.075065 0.000000 0.591377 0.000000 -0.380313 0.000000 ) omega( 3) = 0.089782 [THz] = 2.994825 [cm-1] ( -0.565099 -0.000000 -0.176836 -0.000000 -0.386513 -0.000000 ) ( -0.565099 -0.000000 -0.176836 -0.000000 -0.386513 -0.000000 ) omega( 4) = 15.294047 [THz] = 510.157877 [cm-1] ( 0.489220 0.000000 -0.455114 0.000000 -0.231375 0.000000 ) ( -0.489220 0.000000 0.455114 0.000000 0.231375 0.000000 ) omega( 5) = 15.294047 [THz] = 510.157877 [cm-1] ( 0.481628 -0.000000 0.517721 -0.000000 -0.000000 -0.000000 ) ( -0.481628 -0.000000 -0.517721 -0.000000 -0.000000 -0.000000 ) omega( 6) = 15.294047 [THz] = 510.157877 [cm-1] ( -0.169406 0.000000 0.157595 0.000000 -0.668181 0.000000 ) ( 0.169406 0.000000 -0.157595 0.000000 0.668181 0.000000 ) ************************************************************************** -- ___________________________________________________ Cristian Degli Esposti Boschi CNR, CNISM, Unita' di Ricerca di Bologna, c/o Dipartimento di Fisica, Universita' di Bologna viale Berti-Pichat, 6/2, 40127, Bologna, Italia tel. ++39 051 2095114 fax ++39 051 2095113 e-mail: cristian.degliesposti -AT- unibo.it web: http://www.df.unibo.it/fismat/theory ___________________________________________________ From akohlmey at cmm.chem.upenn.edu Fri May 8 21:06:22 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 08 May 2009 15:06:22 -0400 Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: <1241809582.2961.398.camel@zero> On Fri, 2009-05-08 at 12:09 -0500, Karoly Nemeth wrote: > Thanks Y. C. Cheng! > > However, I do not see any listing of the 4.1 CVS version from the PWSCF > download page or elsewhere. How can I get it? through: http://www.qe-forge.it/ you can either access the latest version via cvs or download nightly snapshots. cheers, axel. > Thanks: > > Karoly > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Fri, 8 May 2009 10:50:54 +0800 > > From: ??? > > Subject: Re: [Pw_forum] QE 4.0.5 pseudopotencials > > To: PWSCF Forum > > Message-ID: > > > > Content-Type: text/plain; charset="utf-8" > > > > Dear karoly, > > You can use ld1.x in 4.1CVS to generate UPF. > > > > Y. C. Cheng > > Department of Physics > > Nanjing University > > Nanjing 210093 > > P. R. China > > Tel: 86-25-83592907 > > Email: yccheng.nju at gmail.com > > -------------- next part -------------- > > An HTML attachment was scrubbed... > > URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/8d23be4d/attachment.html > > > > ------------------------------ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > End of Pw_forum Digest, Vol 23, Issue 19 > > **************************************** > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From degironc at sissa.it Fri May 8 21:47:03 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 08 May 2009 21:47:03 +0200 Subject: [Pw_forum] Format of dynamical matrices In-Reply-To: <4A046B1A.5050809@unibo.it> References: <4A046B1A.5050809@unibo.it> Message-ID: <4A048C37.4050300@sissa.it> Cristian Degli Esposti Boschi wrote: > (1) After the line "Dynamical Matrix in cartesian axes" > and the q-value (0 0 0 in my cases) I see a block > of 3 rows and 6 columns for each possible choice of say (n,m) > where n and m label the atoms. What do the entries of > the rows and columns represent? It seems that the second > element of every entry is always 0... > it'n not a 3x6 matrix but a 3x3 complx matrix of second order derivatives w.r.t. atomic dispacements > (2) After the line "Diagonalizing the dynamical matrix" and > the q-value, enclosed in two lines of ***** I see the dynamical > matrix expressed as eigenvalues/eigenvectors. > Are the components of the latter expressed (for every atom) in real > space? Again what are the zeroes after each component? > > the eigenvector are 3Nat dimesional complex vectors. one line per atom... HTH stefano From baroni at sissa.it Fri May 8 22:03:23 2009 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 8 May 2009 22:03:23 +0200 Subject: [Pw_forum] dos In-Reply-To: <1241798154.2961.78.camel@zero> References: <297190.3959.qm@web54404.mail.yahoo.com> <9873DED2-6DDC-448B-9045-8287DD1B93A0@sissa.it> <49FFEA2A.7020908@sissa.it> <157F434F-CFC7-4F87-A043-2D346BC17671@sissa.it> <4A015A92.7010408@cea.fr> <47431d9f0905080520p1fb999e2ha72850c6ff18352a@mail.gmail.com> <1241787081.3356.8.camel@zero> <7632F40A-DA5A-4C66-AA80-42B26F3E3DAB@sissa.it> <1241798154.2961.78.camel@zero> Message-ID: On May 8, 2009, at 5:55 PM, Axel Kohlmeyer wrote: > particularly the notion of anti-bonding MOs worries me. those tend > to have restricted meaning in pseudopotential DFT calculations to > begin with. well, it depends. I guess that the antibonding combination of atomic arbitals that ar occupied in the atomic ground state is not any worse than the lbonding combination ... surely, linear combinations of atomic orbitals that are not occupied in the atomic ground state begome less and less meaningful as the enrgy grows large. > in any case one has to be well aware of the fact, that the molecular > wavefunction that the DOS are projected on is to a significant degree > arbitrarily chosen and it is based on an assumption of non-interacting > molecules. not differently of how atomic pdos are based on the assumption of non- interacting atoms. actually, it is preciseley for assessing the deviation from a non-interacting atom situation that the atomic pdos are most useful ... creers - S. --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090508/5b3713e7/attachment.htm From carlo.nervi at unito.it Sat May 9 08:59:48 2009 From: carlo.nervi at unito.it (Carlo Nervi) Date: Sat, 9 May 2009 08:59:48 +0200 (CEST) Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: <1241809582.2961.398.camel@zero> References: <1241809582.2961.398.camel@zero> Message-ID: <3606.151.62.50.73.1241852388.squirrel@mail.unito.it> Hello, maybe i cannot find what's in front of my nose, but I was unable to find the download link of the CVS dev version through: http://www.qe-forge.it/ Is it necessary to login in order to download the v.4.1? Cheers, Carlo -- ------------------------------------------------------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ > > > ------------------ Messaggio originale ------------------- > Oggetto: Re: [Pw_forum] QE 4.0.5 pseudopotencials > Da: "Axel Kohlmeyer" > Data: Ven, 8 Maggio 2009, 9:06 pm > A: "PWSCF Forum" > ---------------------------------------------------------- > > On Fri, 2009-05-08 at 12:09 -0500, Karoly Nemeth wrote: >> Thanks Y. C. Cheng! >> >> However, I do not see any listing of the 4.1 CVS version >> from the PWSCF >> download page or elsewhere. How can I get it? > > through: http://www.qe-forge.it/ > you can either access the latest version via cvs > or download nightly snapshots. > > cheers, > axel. > >> Thanks: >> >> Karoly >> >> >> > ---------------------------------------------------------------------- >> > >> > Message: 1 >> > Date: Fri, 8 May 2009 10:50:54 +0800 >> > From: ??? >> > Subject: Re: [Pw_forum] QE 4.0.5 pseudopotencials >> > To: PWSCF Forum >> > Message-ID: >> > >> > Content-Type: text/plain; charset="utf-8" >> > >> > Dear karoly, >> > You can use ld1.x in 4.1CVS to generate UPF. >> > >> > Y. C. Cheng >> > Department of Physics >> > Nanjing University >> > Nanjing 210093 >> > P. R. China >> > Tel: 86-25-83592907 >> > Email: yccheng.nju at gmail.com >> > -------------- next part -------------- >> > An HTML attachment was scrubbed... >> > URL: >> http://www.democritos.it/pipermail/pw_forum/attachments/20090508/8d23be4d/attachment.html >> > >> > ------------------------------ >> > >> > _______________________________________________ >> > Pw_forum mailing list >> > Pw_forum at pwscf.org >> > http://www.democritos.it/mailman/listinfo/pw_forum >> > >> > >> > End of Pw_forum Digest, Vol 23, Issue 19 >> > **************************************** >> > >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of > Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, > PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: > 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a > better idiot. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > From degironc at sissa.it Sat May 9 10:27:35 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sat, 09 May 2009 10:27:35 +0200 Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: <3606.151.62.50.73.1241852388.squirrel@mail.unito.it> References: <1241809582.2961.398.camel@zero> <3606.151.62.50.73.1241852388.squirrel@mail.unito.it> Message-ID: <4A053E77.1030609@sissa.it> go to http://www.qe-forge.org choose Quantum ESPRESSO in the list on your right click on SCM (I guess it stands for Source Control Manager) and follow the instructions in that page or just click on the link at the bottom. (The night snapshot of today fails to compile PP/pw2gw.f90 for a stupid declaration error that has just been corrected in the CVS repository) HTH stefano Carlo Nervi wrote: > Hello, > maybe i cannot find what's in front of my nose, but I was > unable to find the download link of the CVS dev version > through: http://www.qe-forge.it/ > Is it necessary to login in order to download the v.4.1? > Cheers, > Carlo > -- > ------------------------------------------------------ > Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 > Fax: +39 011 6707855 - Dipartimento di Chimica IFM > via P. Giuria 7, 10125 Torino, Italy > http://lem.ch.unito.it/ > > >> ------------------ Messaggio originale ------------------- >> Oggetto: Re: [Pw_forum] QE 4.0.5 pseudopotencials >> Da: "Axel Kohlmeyer" >> Data: Ven, 8 Maggio 2009, 9:06 pm >> A: "PWSCF Forum" >> ---------------------------------------------------------- >> >> On Fri, 2009-05-08 at 12:09 -0500, Karoly Nemeth wrote: >> >>> Thanks Y. C. Cheng! >>> >>> However, I do not see any listing of the 4.1 CVS version >>> from the PWSCF >>> download page or elsewhere. How can I get it? >>> >> through: http://www.qe-forge.it/ >> you can either access the latest version via cvs >> or download nightly snapshots. >> >> cheers, >> axel. >> >> >>> Thanks: >>> >>> Karoly >>> >>> >>> >>>> ---------------------------------------------------------------------- >>>> >>>> Message: 1 >>>> Date: Fri, 8 May 2009 10:50:54 +0800 >>>> From: ??? >>>> Subject: Re: [Pw_forum] QE 4.0.5 pseudopotencials >>>> To: PWSCF Forum >>>> Message-ID: >>>> >>>> Content-Type: text/plain; charset="utf-8" >>>> >>>> Dear karoly, >>>> You can use ld1.x in 4.1CVS to generate UPF. >>>> >>>> Y. C. Cheng >>>> Department of Physics >>>> Nanjing University >>>> Nanjing 210093 >>>> P. R. China >>>> Tel: 86-25-83592907 >>>> Email: yccheng.nju at gmail.com >>>> -------------- next part -------------- >>>> An HTML attachment was scrubbed... >>>> URL: >>>> >>> http://www.democritos.it/pipermail/pw_forum/attachments/20090508/8d23be4d/attachment.html >>> >>>> ------------------------------ >>>> >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>>> >>>> End of Pw_forum Digest, Vol 23, Issue 19 >>>> **************************************** >>>> >>>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> -- >> ======================================================================= >> Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu >> http://www.cmm.upenn.edu >> Center for Molecular Modeling -- University of >> Pennsylvania >> Department of Chemistry, 231 S.34th Street, Philadelphia, >> PA 19104-6323 >> tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: >> 1-215-898-5425 >> ======================================================================= >> If you make something idiot-proof, the universe creates a >> better idiot. >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> >> > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From m.abbasnejad at gmail.com Sat May 9 15:19:28 2009 From: m.abbasnejad at gmail.com (mohaddeseh abbasnejad) Date: Sat, 9 May 2009 17:49:28 +0430 Subject: [Pw_forum] periodic boundary condition Message-ID: <7699a1950905090619u1880a361pe371406905804ec5@mail.gmail.com> Dear all, I want to calculate the total energy of Si atom using CASINO without imposing periodic boundary conditions. At the first step, I should prepare a file containing wavefunction's cofficients by QE code. I know that for calculating the total energy of an atom, I have to use supercell, which it also imposes periodic boundary conditions. Would you please guide me how I should prepare the input file for QE such that it doesn't have any periodic boundary condition? Thanks in advance. Yours, Mohaddeseh -- --------------------------------------------------------- Mohaddeseh Abbasnejad, Room No. 323, Department of Physics, University of Tehran, North Karegar Ave., Tehran, P.O. Box: 14395-547- IRAN Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 Cellphone: +989177317514 E-Mail: m.abbasnejad at gmail.com Website: http://physics.ut.ac.ir --------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090509/58753c33/attachment.htm From akohlmey at cmm.chem.upenn.edu Sat May 9 15:47:09 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 09 May 2009 09:47:09 -0400 Subject: [Pw_forum] periodic boundary condition In-Reply-To: <7699a1950905090619u1880a361pe371406905804ec5@mail.gmail.com> References: <7699a1950905090619u1880a361pe371406905804ec5@mail.gmail.com> Message-ID: <1241876829.3313.15.camel@zero> On Sat, 2009-05-09 at 17:49 +0430, mohaddeseh abbasnejad wrote: > Dear all, > > I want to calculate the total energy of Si atom using CASINO without > imposing periodic boundary conditions. > At the first step, I should prepare a file containing wavefunction's > cofficients by QE code. > I know that for calculating the total energy of an atom, I have to use > supercell, which it also imposes periodic boundary conditions. > Would you please guide me how I should prepare the input file for QE > such that it doesn't have any periodic boundary condition? mohaddeseh, Q-E uses plane waves as basis functions and those are by construction periodic. you can approximate the energy of a single atom by systematically increasing the cell size and extrapolating to infinity. however, if you pay attention to the relative error vs. infinite cell and compare it e.g., to the intrinsic error of DFT or the error of using a finite basis set etc., then you'll be able to determine a cell size where the error from the periodicity (or limited cell size) is sufficiently smaller than the others and thus it would not matter that you have periodic boundaries. that all being said, wouldn't it be easier to use a code that utilizes a localized basis set for a problem like this? as little as i know about QMC code, they usually support some of those as well. cheers, axel. > Thanks in advance. > > Yours, > Mohaddeseh > > > > > -- > --------------------------------------------------------- > > Mohaddeseh Abbasnejad, > Room No. 323, Department of Physics, > University of Tehran, North Karegar Ave., > Tehran, P.O. Box: 14395-547- IRAN > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > Cellphone: +989177317514 > E-Mail: m.abbasnejad at gmail.com > Website: http://physics.ut.ac.ir > > --------------------------------------------------------- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From lex at phys.ufl.edu Sat May 9 19:59:44 2009 From: lex at phys.ufl.edu (Lex Kemper) Date: Sat, 09 May 2009 13:59:44 -0400 Subject: [Pw_forum] periodic boundary condition In-Reply-To: <7699a1950905090619u1880a361pe371406905804ec5@mail.gmail.com> References: <7699a1950905090619u1880a361pe371406905804ec5@mail.gmail.com> Message-ID: <4A05C490.1080106@phys.ufl.edu> Hi Mohaddeseh, As far as I know what you're asking for is not possible. The best you can do if place the single atom in a large box, taking care to make sure that (a) the box is large enough and (b) your energy cutoff is large enough. Also, take care with the Si's spin state. Cheers, Lex Kemper University of Florida mohaddeseh abbasnejad wrote: > Dear all, > > I want to calculate the total energy of Si atom using CASINO without > imposing periodic boundary conditions. > At the first step, I should prepare a file containing wavefunction's > cofficients by QE code. > I know that for calculating the total energy of an atom, I have to use > supercell, which it also imposes periodic boundary conditions. > Would you please guide me how I should prepare the input file for QE > such that it doesn't have any periodic boundary condition? > Thanks in advance. > > Yours, > Mohaddeseh > > > > -- > --------------------------------------------------------- > > Mohaddeseh Abbasnejad, > Room No. 323, Department of Physics, > University of Tehran, North Karegar Ave., > Tehran, P.O. Box: 14395-547- IRAN > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > Cellphone: +989177317514 > E-Mail: m.abbasnejad at gmail.com > Website: http://physics.ut.ac.ir > > --------------------------------------------------------- > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From nemeth at anl.gov Sun May 10 05:33:27 2009 From: nemeth at anl.gov (Karoly Nemeth) Date: Sat, 9 May 2009 22:33:27 -0500 (CDT) Subject: [Pw_forum] Pw_forum Digest, Vol 23, Issue 23 In-Reply-To: References: Message-ID: Thank you Axel! Karoly >> However, I do not see any listing of the 4.1 CVS version from the PWSCF >> download page or elsewhere. How can I get it? > > through: http://www.qe-forge.it/ > you can either access the latest version via cvs > or download nightly snapshots. > > cheers, > axel. > From tunderaji at gmail.com Sun May 10 20:04:41 2009 From: tunderaji at gmail.com (Raji Abdulrafiu) Date: Sun, 10 May 2009 20:04:41 +0200 Subject: [Pw_forum] Format of dynamical matrices Message-ID: <306556ed0905101104w76d2d2data368b170152a34d4@mail.gmail.com> Thanks Stefano. However, personally, I am not yet clear about your explanation regarding the dynamical matrices. Please explain in clear terms what those zeros in second , fourth and sixth columns after the 'Dynamical Matrix in Cartesian axes' stand for. Also the zeros in similar columns after the "Diagonalizing the dynamical matrix. What do they correspond to? Thank you very much. -- Raji Abdulrafiu Tunde, Solid state and Material Physics Group, Department of Physics, University of Cape Town, Private Bag X3, Rondebosch 7701, Cape Town. South Africa. Cell: +27722668228 Fax: +27216503342 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090510/b857efa1/attachment.htm From degironc at sissa.it Sun May 10 20:55:05 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Sun, 10 May 2009 20:55:05 +0200 Subject: [Pw_forum] Format of dynamical matrices In-Reply-To: <306556ed0905101104w76d2d2data368b170152a34d4@mail.gmail.com> References: <306556ed0905101104w76d2d2data368b170152a34d4@mail.gmail.com> Message-ID: <4A072309.60404@sissa.it> this means that your matrix is real stefano Raji Abdulrafiu wrote: > Thanks Stefano. However, personally, I am not yet clear about your > explanation regarding the dynamical matrices. Please explain in clear > terms what those zeros in second , fourth and sixth columns after > the 'Dynamical Matrix in Cartesian axes' stand for. Also the zeros in > similar columns after the "Diagonalizing the dynamical matrix. What do > they correspond to? > > Thank you very much. > -- > Raji Abdulrafiu Tunde, > Solid state and Material Physics Group, > Department of Physics, > University of Cape Town, > Private Bag X3, > Rondebosch 7701, > Cape Town. South Africa. > Cell: +27722668228 > Fax: +27216503342 > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From m.abbasnejad at gmail.com Mon May 11 11:07:15 2009 From: m.abbasnejad at gmail.com (mohaddeseh abbasnejad) Date: Mon, 11 May 2009 13:37:15 +0430 Subject: [Pw_forum] ion relaxation Message-ID: <7699a1950905110207h5378e6f2n87241a490623c99f@mail.gmail.com> Dear all, After doing ion relaxation for Si, I received such an error and the program stopped: "convergence NOT achieved after 100 iterations: stopping" Would you please help me? Thanks in advance. Yours, Mohaddeseh -- --------------------------------------------------------- Mohaddeseh Abbasnejad, Room No. 323, Department of Physics, University of Tehran, North Karegar Ave., Tehran, P.O. Box: 14395-547- IRAN Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 Cellphone: +989177317514 E-Mail: m.abbasnejad at gmail.com Website: http://physics.ut.ac.ir --------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090511/f4eddb6f/attachment.html From Giovanni.Cantele at na.infn.it Mon May 11 11:38:57 2009 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Mon, 11 May 2009 11:38:57 +0200 Subject: [Pw_forum] ion relaxation In-Reply-To: <7699a1950905110207h5378e6f2n87241a490623c99f@mail.gmail.com> References: <7699a1950905110207h5378e6f2n87241a490623c99f@mail.gmail.com> Message-ID: <4A07F231.6020708@na.infn.it> mohaddeseh abbasnejad wrote: > Dear all, > > After doing ion relaxation for Si, I received such an error and the > program stopped: > "convergence NOT achieved after 100 iterations: stopping" > Would you please help me? > Thanks in advance. > > Yours, > Mohaddeseh > > -- > --------------------------------------------------------- > > Mohaddeseh Abbasnejad, > Room No. 323, Department of Physics, > University of Tehran, North Karegar Ave., > Tehran, P.O. Box: 14395-547- IRAN > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > Cellphone: +989177317514 > E-Mail: m.abbasnejad at gmail.com > Website: http://physics.ut.ac.ir > > --------------------------------------------------------- > By default the relaxation run stops either if convergence has been achieved or if the total number of ionic + electronics steps exceeds nstep. Usually, the default nstep is sufficient to obtain convergence on ion relaxation. What to do? i) try to increase nstep, just to see if your system needs more steps to converge (possibly, you can restart your previous run, so you'll not loose what done until now!) ii) check your starting config., if "too unphysical" or far from the final configuration the code might require a very large number of steps to converge iii) more hints in the link below: http://www.quantum-espresso.org/wiki/index.php/Troubleshooting_(PWscf)#structural_optimization_is_slow_or_does_not_converge Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it Skype contact: giocan74 From sclauzer at sissa.it Mon May 11 11:44:47 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 11 May 2009 11:44:47 +0200 Subject: [Pw_forum] ion relaxation In-Reply-To: <7699a1950905110207h5378e6f2n87241a490623c99f@mail.gmail.com> References: <7699a1950905110207h5378e6f2n87241a490623c99f@mail.gmail.com> Message-ID: <4A07F38F.1080404@sissa.it> Dear Mohaddeseh, first read here http://www.quantum-espresso.org/wiki/index.php/Frequently_Asked_Questions#Self-consistency_is_slow_or_does_not_converge_at_all and related questions. Then, if you cannot still figure out anything you can write again. Gabriele mohaddeseh abbasnejad wrote: > Dear all, > > After doing ion relaxation for Si, I received such an error and the > program stopped: > "convergence NOT achieved after 100 iterations: stopping" > Would you please help me? > Thanks in advance. > > Yours, > Mohaddeseh > > -- > --------------------------------------------------------- > > Mohaddeseh Abbasnejad, > Room No. 323, Department of Physics, > University of Tehran, North Karegar Ave., > Tehran, P.O. Box: 14395-547- IRAN > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > Cellphone: +989177317514 > E-Mail: m.abbasnejad at gmail.com > Website: http://physics.ut.ac.ir > > --------------------------------------------------------- > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From cristian.degliesposti at unibo.it Mon May 11 12:00:54 2009 From: cristian.degliesposti at unibo.it (Cristian Degli Esposti Boschi) Date: Mon, 11 May 2009 12:00:54 +0200 Subject: [Pw_forum] Format of dynamical matrices Message-ID: <4A07F756.8000500@unibo.it> Thanks for the answer regarding the format of the dynamical matrix. Are the eigenvectors expressed in real space? What are the units in this case? -- ___________________________________________________ Cristian Degli Esposti Boschi CNR, CNISM, Unita' di Ricerca di Bologna, c/o Dipartimento di Fisica, Universita' di Bologna viale Berti-Pichat, 6/2, 40127, Bologna, Italia tel. ++39 051 2095114 fax ++39 051 2095113 e-mail: cristian.degliesposti -AT- unibo.it web: http://www.df.unibo.it/fismat/theory ___________________________________________________ From hqzhou at nju.edu.cn Mon May 11 12:12:54 2009 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Mon, 11 May 2009 18:12:54 +0800 Subject: [Pw_forum] Using q point parallelization Message-ID: <67A6A0EE4042441AB22B7CB4B8402144@solarflare> Hi, list users, I'm trying the manual q-point parallelization and have one question to confirm. In the additional information section of INPUT_PH.html of 4.1 cvs, it indicates that after running different subset of irrs or qs on each machine, one need to "collect all the data-file.xml.#ip.#irr files in one directory and run ph.x." My question is why one needs to run ph.x again instead of using q2r directly to calculate FC. If this is neccessary, what changes should be made in the input file. Thanks in advance! Huiqun Zhou @Earth Sciences, Nanjing University, China From xylnew at 163.com Tue May 12 01:15:35 2009 From: xylnew at 163.com (xylnew) Date: Tue, 12 May 2009 07:15:35 +0800 (CST) Subject: [Pw_forum] how to control Berry Phase convergence Message-ID: <25786673.529501242083735668.JavaMail.coremail@bj163app15.163.com> Dear all, I want to do some Berry Phase calculations with CP code in QE. Are the covnergence parameters different from the ones of other type calculations(total energy and normal cp etc.)? Thanks. Xieyl www.syiae.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090512/918afcad/attachment.htm From nemeth at anl.gov Tue May 12 02:55:34 2009 From: nemeth at anl.gov (Karoly Nemeth) Date: Mon, 11 May 2009 19:55:34 -0500 (CDT) Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: Hi, I have downloaded q-e-scm-2009-05-10 and generated some new pseudopotentials in UPF format. I tried to run them with QE 4.0.4. I got from read_upf_v2 : error # 1 Unknown UPF format version: 2.0.1 . Can I use the new PPs with the 4.1 series only? Can I use the latest FFTW with the 4.1 series with q-e-scm-2009-05-10, or should I continue using the older 2.1.5 series ? Thanks: Karoly *********************************************************************** Karoly Nemeth, Ph.D. Argonne National Laboratory 9700 S. Cass Ave. Argonne, IL 60439. Advanced Photon Source X-Ray Science Division Building 401, Room B4198 Tel. 630-252-5813 Fax: 630-252-3222 *********************************************************************** From jasius_1 at yahoo.com Tue May 12 03:14:15 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Mon, 11 May 2009 18:14:15 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <25786673.529501242083735668.JavaMail.coremail@bj163app15.163.com> Message-ID: <961443.30084.qm@web54406.mail.yahoo.com> I found discrepancies in how monoclinic cell is defined in the manual and how it's defined in experimental crystalography files: Namely, For P lattices: the special axis (c) is the z-axis, one basal-plane vector (a) is along x, the other basal-plane vector (b) is at angle gamma for monoclinic. While it is not clear angle to which axis, later on gamma is defined as an angle between a and b. the problem is that ab is 90 degree angle, not a unique angle, if I visualize the structure with cell axes. Furthermore, when I open any crystalography file in Mercury, I get for monoclinic cell alpha 90.00 betta 115.6670(10) gamma 90.00 so gamma is again not a unique angle but rather 90 degree angle. I assume this is not what needs to be entered for monoclinic lattice in celldm(4), but rather a unique angle which in both cases is between a and c, so it has nothing to do with b as defined above Could anybody help out? Jonas Baltrusaitis University of Iowa From zqwu00 at gmail.com Tue May 12 03:17:07 2009 From: zqwu00 at gmail.com (zq wu) Date: Mon, 11 May 2009 18:17:07 -0700 Subject: [Pw_forum] Using q point parallelization In-Reply-To: <67A6A0EE4042441AB22B7CB4B8402144@solarflare> References: <67A6A0EE4042441AB22B7CB4B8402144@solarflare> Message-ID: <3ebf88c60905111817q23f43e58ld2af4651a74e8987@mail.gmail.com> On Mon, May 11, 2009 at 3:12 AM, Huiqun Zhou wrote: > Hi, list users, > > I'm trying the manual q-point parallelization and have one > question to confirm. In the additional information section > of INPUT_PH.html of 4.1 cvs, it indicates that after running > different subset of irrs or qs on each machine, one need to > "collect all the data-file.xml.#ip.#irr files in one directory > and run ph.x." My question is why one needs to run ph.x again > instead of using q2r directly to calculate FC. If this is > neccessary, what changes should be made in the input file. To get the dyn files, which is input of q2r to calculate FC. You do not need change the input file except specifying "recovery=.true." > > > Thanks in advance! > > Huiqun Zhou > @Earth Sciences, Nanjing University, China > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > Zhongqing CACS University of Southern California -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090511/4837844f/attachment.htm From jasius_1 at yahoo.com Tue May 12 03:51:34 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Mon, 11 May 2009 18:51:34 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question Message-ID: <402489.1178.qm@web54410.mail.yahoo.com> To reiterate on my previous question: while a,b,c parameters can be transferred to pwscf unambiguously, angles ab, bc and ac are not ac clear. Are they equivalent to alfa beta and gamma, respectively, according to Wyckoff definition? Jonas --- On Mon, 5/11/09, Jonas Baltrusaitis wrote: > From: Jonas Baltrusaitis > Subject: monoclinic cell definition question > To: "PWSCF Forum" > Date: Monday, May 11, 2009, 6:14 PM > I found discrepancies in how monoclinic cell is defined in > the manual and how it's defined in experimental > crystalography files: > > Namely, > > For P lattices: the special axis (c) is the z-axis, one > basal-plane > vector (a) is along x, the other basal-plane vector (b) is > at angle > gamma for monoclinic. While it is not clear angle to which > axis, later on gamma is defined as an angle between a and b. > the problem is that ab is 90 degree angle, not a unique > angle, if I visualize the structure with cell axes. > > Furthermore, when I open any crystalography file in > Mercury, I get for monoclinic cell > > alpha 90.00 betta 115.6670(10) gamma 90.00 so gamma is > again not a unique angle but rather 90 degree angle. I > assume this is not what needs to be entered for monoclinic > lattice in celldm(4), but rather a unique angle which in > both cases is between a and c, so it has nothing to do with > b as defined above > > Could anybody help out? > > Jonas Baltrusaitis > University of Iowa From sclauzer at sissa.it Tue May 12 09:10:46 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 12 May 2009 09:10:46 +0200 Subject: [Pw_forum] Using q point parallelization In-Reply-To: <3ebf88c60905111817q23f43e58ld2af4651a74e8987@mail.gmail.com> References: <67A6A0EE4042441AB22B7CB4B8402144@solarflare> <3ebf88c60905111817q23f43e58ld2af4651a74e8987@mail.gmail.com> Message-ID: <4A0920F6.6090907@sissa.it> zq wu wrote: > > > On Mon, May 11, 2009 at 3:12 AM, Huiqun Zhou > wrote: > > Hi, list users, > > I'm trying the manual q-point parallelization and have one > question to confirm. In the additional information section > of INPUT_PH.html of 4.1 cvs, it indicates that after running > different subset of irrs or qs on each machine, one need to > "collect all the data-file.xml.#ip.#irr files in one directory > and run ph.x." My question is why one needs to run ph.x again > instead of using q2r directly to calculate FC. Because in this additional run ph.x will collect the data produced by all previous runs to build the full dynamical matrix and write it to the files (one for each q-point) with the prefix specified in fildyn. These files are needed by q2r to compute the interatomic force constants. I think that those files should not be produced by ph.x after partial runs (e.g. with only a subset of irreps) in order to avoid confusion, but maybe this was fixed in a later version of the CVS. > If this is > neccessary, what changes should be made in the input file. > > To get the dyn files, which is input of q2r to calculate FC. > You do not need change the input file except specifying > "recovery=.true." recover=.true. GS > > > > Thanks in advance! > > Huiqun Zhou > @Earth Sciences, Nanjing University, China > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > Zhongqing > > CACS University of Southern California > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From ddrsmile at gmail.com Tue May 12 09:27:02 2009 From: ddrsmile at gmail.com (DdrSmiLe) Date: Tue, 12 May 2009 15:27:02 +0800 Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <998756.50141.qm@web65701.mail.ac4.yahoo.com> References: <4A0182D1.7010805@cam.ac.uk> <998756.50141.qm@web65701.mail.ac4.yahoo.com> Message-ID: Hi Eyvaz lsaev I've tried your files. Thank you for your sharing. I have some question as following. Would your code print out the T_c? If so, where would it be printed? if not, what's the extra procedure? Sorry for asking such basic questions. My major is not chemistry or physics, but I'm doing job related to them Therefore, there are too many things confuse me. Thank you for your helping Joey, Liu 2009/5/7 Eyvaz Isaev > Hi, > > Sorry for empty mail, unexpectedly I clicked on Send button. > > Well, I think attached files are what you need. The first one (modified > elphon.f90) writes special files, and the second one uses these files to > find lambda and then T_c via the Allen-Dynes equation (\omega_log instead of > the Debye temperature). > > In fact, recently I suggested to change original elphon.f90 and lambda.f90 > files with those (at least, include changes), but I do not know whether this > suggestion was accepted by developers. > Attached elphon.f90 worked for QE 4.0.4, at least. > > Any questions, please contact me. > > Bests, > Eyvaz. > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, > Russia, > Department of Physics, Chemistry, and Biology (IFM), Linkoping University, > Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com > > > --- On Wed, 5/6/09, David Tompsett wrote: > > > From: David Tompsett > > Subject: [Pw_forum] Documentation on El-Ph coupling > > To: pw_forum at pwscf.org > > Date: Wednesday, May 6, 2009, 4:30 PM > > Dear All, > > I am interested in calculating the > > electron-phonon coupling > > of a metallic system in QE. I would like to then apply the > > McMillan > > formula for the prediction of the superconducting Tc. > > > > In the basic tutorials list, /Hands-on Tutorial of Quantum > > Espresso > > > >,/ > > I could find > > examples on the El-ph calculation for Aluminium at X. Is > > there any > > further documentation available to indicated on how to find > > the > > integrated lambda and wlog that enter the McMillan formula > > for Tc? > > > > Thank you, > > David. > > > > -- > > David A. Tompsett > > Quantum Matter Group > > Cavendish Laboratory > > J. J. Thomson Avenue > > Cambridge CB3 0HE > > U.K. > > Tel: +44 7907 566351 (mobile) > > Fax: +44 1223 768140 > > http://www-qm.phy.cam.ac.uk/ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090512/c775b20a/attachment.htm From xenia at itim-cj.ro Tue May 12 10:36:43 2009 From: xenia at itim-cj.ro (Xenia Filip) Date: Tue, 12 May 2009 11:36:43 +0300 (EEST) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <961443.30084.qm@web54406.mail.yahoo.com> References: <961443.30084.qm@web54406.mail.yahoo.com> Message-ID: <1242.193.231.25.43.1242117403.squirrel@webmail.itim-cj.ro> I have the same problem recently and due to a monoclinic cell I realize that I introduce in a wrong manner the coordinates of atoms and cell parameters in the input file. I admit that I didn't read carefully the INPUT_PW.txt file. Due to the discrepancy you talk I think that is not correct to take (x,y,z), (a,b,c) and (alpha, beta, gamma) from a *.cif file and put them on the same order in *scf.in file. I talk about the convention used in *.cif files with somebody which is doing X Ray Crystalography and if he has right the "b" axis in *.cif I think it correspond to "c" axis in *.scf.in. That means that it needs to permute (a,b,c) in (c,a,b), (x,y,z) in (z,x,y), (alpha, beta, gamma) in (gamma, alpha, beta). I am not expert in doing such calculation, I am interest just in chemical shift calculations (I'm doing NMR), and unfortunatelly I didn't succed yet to verify this. I am running now some test but it take some time (I do not have strong computers). I'm not sure if my ideea is correct but you can think at this posibility too. Dr. Xenia Filip National Institute for R&D of Isotopic and Molecular Technologies Cluj 400293, Romania Tel: 0040 724 321420 Fax: 0040 264 420042 > > I found discrepancies in how monoclinic cell is defined in the manual and > how it's defined in experimental crystalography files: > > Namely, > > For P lattices: the special axis (c) is the z-axis, one basal-plane > vector (a) is along x, the other basal-plane vector (b) is at angle > gamma for monoclinic. While it is not clear angle to which axis, later on > gamma is defined as an angle between a and b. the problem is that ab is 90 > degree angle, not a unique angle, if I visualize the structure with cell > axes. > > Furthermore, when I open any crystalography file in Mercury, I get for > monoclinic cell > > alpha 90.00 betta 115.6670(10) gamma 90.00 so gamma is again not a unique > angle but rather 90 degree angle. I assume this is not what needs to be > entered for monoclinic lattice in celldm(4), but rather a unique angle > which in both cases is between a and c, so it has nothing to do with b as > defined above > > Could anybody help out? > > Jonas Baltrusaitis > University of Iowa > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From paulatto at sissa.it Tue May 12 10:07:27 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Tue, 12 May 2009 10:07:27 +0200 (CEST) Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: <41956.78.12.161.199.1242115647.squirrel@webmail.sissa.it> On Tue, May 12, 2009 02:55, Karoly Nemeth wrote: > Unknown UPF format version: 2.0.1 . > Can I use the new PPs with the 4.1 series only? In theory yes, but as the only difference is that fixed bug, you can just open the UPF file with a notepad and change version="2.0.1" to version="2.0.0" > Can I use the latest FFTW with the 4.1 series with q-e-scm-2009-05-10, or > should I continue using the older 2.1.5 series ? Yes, you can use FFT 3.x as well as FFT included in ACML and ESSL libraries. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From jasius_1 at yahoo.com Tue May 12 14:12:30 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 12 May 2009 05:12:30 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <1242.193.231.25.43.1242117403.squirrel@webmail.itim-cj.ro> Message-ID: <431285.79631.qm@web54405.mail.yahoo.com> Thanks Xenia I'll try pwgui maybe it automatically recognizes those. Maybe by that time we'll get answer from forum gurus, or better yet some hands on example (the one in tutorials and manuals doesn't count) Jonas --- On Tue, 5/12/09, Xenia Filip wrote: > From: Xenia Filip > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 12, 2009, 1:36 AM > I have the same problem recently and due to a monoclinic > cell I realize > that I introduce in a wrong manner the coordinates of atoms > and cell > parameters in the input file. I admit that I didn't > read carefully the > INPUT_PW.txt file. Due to the discrepancy you talk I think > that is not > correct to take (x,y,z), (a,b,c) and (alpha, beta, gamma) > from a *.cif > file and put them on the same order in *scf.in file. I talk > about the > convention used in *.cif files with somebody which is doing > X Ray > Crystalography and if he has right the "b" axis > in *.cif I think it > correspond to "c" axis in *.scf.in. That means > that it needs to permute > (a,b,c) in (c,a,b), (x,y,z) in (z,x,y), (alpha, beta, > gamma) in (gamma, > alpha, beta). I am not expert in doing such calculation, I > am interest > just in chemical shift calculations (I'm doing NMR), > and unfortunatelly I > didn't succed yet to verify this. I am running now some > test but it take > some time (I do not have strong computers). > I'm not sure if my ideea is correct but you can think > at this posibility too. > > Dr. Xenia Filip > National Institute for R&D of > Isotopic and Molecular Technologies > Cluj 400293, Romania > Tel: 0040 724 321420 > Fax: 0040 264 420042 > > > > I found discrepancies in how monoclinic cell is > defined in the manual and > > how it's defined in experimental crystalography > files: > > > > Namely, > > > > For P lattices: the special axis (c) is the z-axis, > one basal-plane > > vector (a) is along x, the other basal-plane vector > (b) is at angle > > gamma for monoclinic. While it is not clear angle to > which axis, later on > > gamma is defined as an angle between a and b. the > problem is that ab is 90 > > degree angle, not a unique angle, if I visualize the > structure with cell > > axes. > > > > Furthermore, when I open any crystalography file in > Mercury, I get for > > monoclinic cell > > > > alpha 90.00 betta 115.6670(10) gamma 90.00 so gamma is > again not a unique > > angle but rather 90 degree angle. I assume this is not > what needs to be > > entered for monoclinic lattice in celldm(4), but > rather a unique angle > > which in both cases is between a and c, so it has > nothing to do with b as > > defined above > > > > Could anybody help out? > > > > Jonas Baltrusaitis > > University of Iowa > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > From jasius_1 at yahoo.com Tue May 12 15:09:38 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 12 May 2009 06:09:38 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <1242.193.231.25.43.1242117403.squirrel@webmail.itim-cj.ro> Message-ID: <712465.49457.qm@web54401.mail.yahoo.com> I just checked pwgui, it can't read cif files. So this monoclinic and probably triclinic angle question is not clear. In the forum I found lots abaout ibrav=13, but not specifically about the discrepancy in between which axes which angle is. It is clear for now that it's not the same alfa, beta gamma in cif files as in alpha beta gamma in pwscf input Jonas --- On Tue, 5/12/09, Xenia Filip wrote: > From: Xenia Filip > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 12, 2009, 1:36 AM > I have the same problem recently and due to a monoclinic > cell I realize > that I introduce in a wrong manner the coordinates of atoms > and cell > parameters in the input file. I admit that I didn't > read carefully the > INPUT_PW.txt file. Due to the discrepancy you talk I think > that is not > correct to take (x,y,z), (a,b,c) and (alpha, beta, gamma) > from a *.cif > file and put them on the same order in *scf.in file. I talk > about the > convention used in *.cif files with somebody which is doing > X Ray > Crystalography and if he has right the "b" axis > in *.cif I think it > correspond to "c" axis in *.scf.in. That means > that it needs to permute > (a,b,c) in (c,a,b), (x,y,z) in (z,x,y), (alpha, beta, > gamma) in (gamma, > alpha, beta). I am not expert in doing such calculation, I > am interest > just in chemical shift calculations (I'm doing NMR), > and unfortunatelly I > didn't succed yet to verify this. I am running now some > test but it take > some time (I do not have strong computers). > I'm not sure if my ideea is correct but you can think > at this posibility too. > > Dr. Xenia Filip > National Institute for R&D of > Isotopic and Molecular Technologies > Cluj 400293, Romania > Tel: 0040 724 321420 > Fax: 0040 264 420042 > > > > I found discrepancies in how monoclinic cell is > defined in the manual and > > how it's defined in experimental crystalography > files: > > > > Namely, > > > > For P lattices: the special axis (c) is the z-axis, > one basal-plane > > vector (a) is along x, the other basal-plane vector > (b) is at angle > > gamma for monoclinic. While it is not clear angle to > which axis, later on > > gamma is defined as an angle between a and b. the > problem is that ab is 90 > > degree angle, not a unique angle, if I visualize the > structure with cell > > axes. > > > > Furthermore, when I open any crystalography file in > Mercury, I get for > > monoclinic cell > > > > alpha 90.00 betta 115.6670(10) gamma 90.00 so gamma is > again not a unique > > angle but rather 90 degree angle. I assume this is not > what needs to be > > entered for monoclinic lattice in celldm(4), but > rather a unique angle > > which in both cases is between a and c, so it has > nothing to do with b as > > defined above > > > > Could anybody help out? > > > > Jonas Baltrusaitis > > University of Iowa > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > From wangweiphysics at yahoo.com.cn Tue May 12 17:38:50 2009 From: wangweiphysics at yahoo.com.cn (wangweiphysics at yahoo.com.cn) Date: Tue, 12 May 2009 23:38:50 +0800 (CST) Subject: [Pw_forum] no frequency and mode symmetry output in case.phG.out Message-ID: <677454.84524.qm@web15203.mail.cnb.yahoo.com> Dear all, ???????? I run the Gamma phonon prequency?by ph.x. But when I set the "recover" flag to .true. to restart this calculation from a interrupted program, no frequency and mode symmetry output in case.phG.out. ?????????What is the problem? How can I fix it? ???????? Appreciate any help regarding this problem. ? Regards, Wangwei, Southeast University, P.R. China 2009.5.12 ___________________________________________________________ ????????????????? http://card.mail.cn.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090512/3caedc96/attachment-0001.htm From panda.deng.pan at gmail.com Tue May 12 17:56:53 2009 From: panda.deng.pan at gmail.com (=?GB2312?B?xcu1xw==?=) Date: Tue, 12 May 2009 23:56:53 +0800 Subject: [Pw_forum] no frequency and mode symmetry output in case.phG.out In-Reply-To: <677454.84524.qm@web15203.mail.cnb.yahoo.com> References: <677454.84524.qm@web15203.mail.cnb.yahoo.com> Message-ID: I am not sure your problem with the absence of the inputs. Try to set iverbosity=1. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090512/ff198d91/attachment.htm From m.abbasnejad at gmail.com Tue May 12 18:29:35 2009 From: m.abbasnejad at gmail.com (mohaddeseh abbasnejad) Date: Tue, 12 May 2009 09:29:35 -0700 Subject: [Pw_forum] Hartree-Fock calculation Message-ID: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> Dear all, I think that it's possible to do Hartree-Fock calculation using PWSCF but it needs some utilities which should be installed before using. I was wondering if you could help me. Thanks in advance. Yours, Mohaddeseh -- --------------------------------------------------------- Mohaddeseh Abbasnejad, Room No. 323, Department of Physics, University of Tehran, North Karegar Ave., Tehran, P.O. Box: 14395-547- IRAN Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 Cellphone: +989177317514 E-Mail: m.abbasnejad at gmail.com Website: http://physics.ut.ac.ir --------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090512/1269cfaa/attachment.htm From nemeth at anl.gov Tue May 12 19:41:46 2009 From: nemeth at anl.gov (Karoly Nemeth) Date: Tue, 12 May 2009 12:41:46 -0500 (CDT) Subject: [Pw_forum] QE 4.0.5 pseudopotencials In-Reply-To: References: Message-ID: Excellent! Thank you Lorenzo! Karoly > Message: 4 > Date: Tue, 12 May 2009 10:07:27 +0200 (CEST) > From: "Lorenzo Paulatto" > Subject: Re: [Pw_forum] QE 4.0.5 pseudopotencials > To: "PWSCF Forum" > Message-ID: <41956.78.12.161.199.1242115647.squirrel at webmail.sissa.it> > Content-Type: text/plain;charset=iso-8859-1 > > > On Tue, May 12, 2009 02:55, Karoly Nemeth wrote: >> Unknown UPF format version: 2.0.1 . >> Can I use the new PPs with the 4.1 series only? > > In theory yes, but as the only difference is that fixed bug, you can just > open the UPF file with a notepad and change > version="2.0.1" > to > version="2.0.0" > > >> Can I use the latest FFTW with the 4.1 series with q-e-scm-2009-05-10, or >> should I continue using the older 2.1.5 series ? > > Yes, you can use FFT 3.x as well as FFT included in ACML and ESSL libraries. > > regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ From eyvaz_isaev at yahoo.com Tue May 12 19:47:22 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 12 May 2009 10:47:22 -0700 (PDT) Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: Message-ID: <245408.78234.qm@web65703.mail.ac4.yahoo.com> Dear Liu, --- On Tue, 5/12/09, DdrSmiLe wrote: > Would your code print out the T_c? > If so, where would it be printed? Without any doubts, if everything went properly. You can find T_c printed after lambda.x finished. > My major is not chemistry or physics, but I'm doing job > related to them. Therefore, there are too many things confuse me. > Then yo should read some solid state textbooks (Kittel, Ashcroft-Mermin, Zaiman, Ricahrd Martin, Harrison, Kohanoff, Kantorovich, etc.) This will help you not to be confused. Please provide your affiliation when post a question to this forum. Bests, Eyvaz. > Joey, Liu > > 2009/5/7 Eyvaz Isaev > > > Hi, > > > > Sorry for empty mail, unexpectedly I clicked on Send > button. > > > > Well, I think attached files are what you need. The > first one (modified > > elphon.f90) writes special files, and the second one > uses these files to > > find lambda and then T_c via the Allen-Dynes equation > (\omega_log instead of > > the Debye temperature). > > > > In fact, recently I suggested to change original > elphon.f90 and lambda.f90 > > files with those (at least, include changes), but I do > not know whether this > > suggestion was accepted by developers. > > Attached elphon.f90 worked for QE 4.0.4, at least. > > > > Any questions, please contact me. > > > > Bests, > > Eyvaz. > > > > > ------------------------------------------------------------------- > > Prof. Eyvaz Isaev, > > Theoretical Physics Department, Moscow State Institute > of Steel & Alloys, > > Russia, > > Department of Physics, Chemistry, and Biology (IFM), > Linkoping University, > > Sweden > > Condensed Matter Theory Group, Uppsala University, > Sweden > > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, > eyvaz_isaev at yahoo.com > > > > > > --- On Wed, 5/6/09, David Tompsett > wrote: > > > > > From: David Tompsett > > > Subject: [Pw_forum] Documentation on El-Ph > coupling > > > To: pw_forum at pwscf.org > > > Date: Wednesday, May 6, 2009, 4:30 PM > > > Dear All, > > > I am interested in calculating the > > > electron-phonon coupling > > > of a metallic system in QE. I would like to then > apply the > > > McMillan > > > formula for the prediction of the superconducting > Tc. > > > > > > In the basic tutorials list, /Hands-on Tutorial > of Quantum > > > Espresso > > > > > > >,/ > > > I could find > > > examples on the El-ph calculation for Aluminium > at X. Is > > > there any > > > further documentation available to indicated on > how to find > > > the > > > integrated lambda and wlog that enter the > McMillan formula > > > for Tc? > > > > > > Thank you, > > > David. > > > > > > -- > > > David A. Tompsett > > > Quantum Matter Group > > > Cavendish Laboratory > > > J. J. Thomson Avenue > > > Cambridge CB3 0HE > > > U.K. > > > Tel: +44 7907 566351 (mobile) > > > Fax: +44 1223 768140 > > > http://www-qm.phy.cam.ac.uk/ > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From akohlmey at cmm.chem.upenn.edu Tue May 12 19:55:01 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 12 May 2009 13:55:01 -0400 Subject: [Pw_forum] Hartree-Fock calculation In-Reply-To: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> References: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> Message-ID: <1242150901.3135.35.camel@zero> On Tue, 2009-05-12 at 09:29 -0700, mohaddeseh abbasnejad wrote: > Dear all, > > I think that it's possible to do Hartree-Fock calculation using PWSCF > but it needs some utilities which should be installed before using. dear mohaddeseh, please explain where you get this impression from. > I was wondering if you could help me. no. > Thanks in advance. cheers, axel. > Yours, > Mohaddeseh > > > > > -- > --------------------------------------------------------- > > Mohaddeseh Abbasnejad, > Room No. 323, Department of Physics, > University of Tehran, North Karegar Ave., > Tehran, P.O. Box: 14395-547- IRAN > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > Cellphone: +989177317514 > E-Mail: m.abbasnejad at gmail.com > Website: http://physics.ut.ac.ir > > --------------------------------------------------------- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From jasius_1 at yahoo.com Tue May 12 21:14:45 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 12 May 2009 12:14:45 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <1242.193.231.25.43.1242117403.squirrel@webmail.itim-cj.ro> Message-ID: <556671.31919.qm@web54403.mail.yahoo.com> I just went through an example of monoclinic crystal structure http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html Seems like when compared with description in INPUT_PW b should be swapped with c and beta should be swapped with gamma (going from cif file to pwscf input) Jonas --- On Tue, 5/12/09, Xenia Filip wrote: > From: Xenia Filip > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 12, 2009, 1:36 AM > I have the same problem recently and due to a monoclinic > cell I realize > that I introduce in a wrong manner the coordinates of atoms > and cell > parameters in the input file. I admit that I didn't > read carefully the > INPUT_PW.txt file. Due to the discrepancy you talk I think > that is not > correct to take (x,y,z), (a,b,c) and (alpha, beta, gamma) > from a *.cif > file and put them on the same order in *scf.in file. I talk > about the > convention used in *.cif files with somebody which is doing > X Ray > Crystalography and if he has right the "b" axis > in *.cif I think it > correspond to "c" axis in *.scf.in. That means > that it needs to permute > (a,b,c) in (c,a,b), (x,y,z) in (z,x,y), (alpha, beta, > gamma) in (gamma, > alpha, beta). I am not expert in doing such calculation, I > am interest > just in chemical shift calculations (I'm doing NMR), > and unfortunatelly I > didn't succed yet to verify this. I am running now some > test but it take > some time (I do not have strong computers). > I'm not sure if my ideea is correct but you can think > at this posibility too. > > Dr. Xenia Filip > National Institute for R&D of > Isotopic and Molecular Technologies > Cluj 400293, Romania > Tel: 0040 724 321420 > Fax: 0040 264 420042 > > > > I found discrepancies in how monoclinic cell is > defined in the manual and > > how it's defined in experimental crystalography > files: > > > > Namely, > > > > For P lattices: the special axis (c) is the z-axis, > one basal-plane > > vector (a) is along x, the other basal-plane vector > (b) is at angle > > gamma for monoclinic. While it is not clear angle to > which axis, later on > > gamma is defined as an angle between a and b. the > problem is that ab is 90 > > degree angle, not a unique angle, if I visualize the > structure with cell > > axes. > > > > Furthermore, when I open any crystalography file in > Mercury, I get for > > monoclinic cell > > > > alpha 90.00 betta 115.6670(10) gamma 90.00 so gamma is > again not a unique > > angle but rather 90 degree angle. I assume this is not > what needs to be > > entered for monoclinic lattice in celldm(4), but > rather a unique angle > > which in both cases is between a and c, so it has > nothing to do with b as > > defined above > > > > Could anybody help out? > > > > Jonas Baltrusaitis > > University of Iowa > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > From m.abbasnejad at gmail.com Tue May 12 22:08:04 2009 From: m.abbasnejad at gmail.com (mohaddeseh abbasnejad) Date: Wed, 13 May 2009 00:38:04 +0430 Subject: [Pw_forum] Hartree-Fock calculation In-Reply-To: <1242150901.3135.35.camel@zero> References: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> <1242150901.3135.35.camel@zero> Message-ID: <7699a1950905121308j3120e148w90ccf72fe5cd69d3@mail.gmail.com> Dear Axel, Honestly I heard it from one of my colleagues. In order to him, it's under test. He just knew about it and nothing more. Yours, Mohaddeseh On Tue, May 12, 2009 at 10:25 PM, Axel Kohlmeyer < akohlmey at cmm.chem.upenn.edu> wrote: > On Tue, 2009-05-12 at 09:29 -0700, mohaddeseh abbasnejad wrote: > > Dear all, > > > > I think that it's possible to do Hartree-Fock calculation using PWSCF > > but it needs some utilities which should be installed before using. > > dear mohaddeseh, > > please explain where you get this impression from. > > > I was wondering if you could help me. > > no. > > > Thanks in advance. > > cheers, > axel. > > > Yours, > > Mohaddeseh > > > > > > > > > > -- > > --------------------------------------------------------- > > > > Mohaddeseh Abbasnejad, > > Room No. 323, Department of Physics, > > University of Tehran, North Karegar Ave., > > Tehran, P.O. Box: 14395-547- IRAN > > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > > Cellphone: +989177317514 > > E-Mail: m.abbasnejad at gmail.com > > Website: http://physics.ut.ac.ir > > > > --------------------------------------------------------- > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a better idiot. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- --------------------------------------------------------- Mohaddeseh Abbasnejad, Room No. 323, Department of Physics, University of Tehran, North Karegar Ave., Tehran, P.O. Box: 14395-547- IRAN Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 Cellphone: +989177317514 E-Mail: m.abbasnejad at gmail.com Website: http://physics.ut.ac.ir --------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/f0ee349f/attachment-0001.htm From degironc at sissa.it Tue May 12 22:11:51 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 12 May 2009 22:11:51 +0200 Subject: [Pw_forum] Hartree-Fock calculation In-Reply-To: <7699a1950905121308j3120e148w90ccf72fe5cd69d3@mail.gmail.com> References: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> <1242150901.3135.35.camel@zero> <7699a1950905121308j3120e148w90ccf72fe5cd69d3@mail.gmail.com> Message-ID: <4A09D807.8070300@sissa.it> mohaddeseh abbasnejad wrote: > Dear Axel, > > Honestly I heard it from one of my colleagues. In order to him, it's > under test. He just knew about it and nothing more. indeed ... read example/EXX_example/README Stefano de Gironcoli - SISSA and DEMOCRITOS > Yours, > Mohaddeseh > > > > On Tue, May 12, 2009 at 10:25 PM, Axel Kohlmeyer > > wrote: > > On Tue, 2009-05-12 at 09:29 -0700, mohaddeseh abbasnejad wrote: > > Dear all, > > > > I think that it's possible to do Hartree-Fock calculation using > PWSCF > > but it needs some utilities which should be installed before using. > > dear mohaddeseh, > > please explain where you get this impression from. > > > I was wondering if you could help me. > > no. > > > Thanks in advance. > > cheers, > axel. > > > Yours, > > Mohaddeseh > > > > > > > > > > -- > > --------------------------------------------------------- > > > > Mohaddeseh Abbasnejad, > > Room No. 323, Department of Physics, > > University of Tehran, North Karegar Ave., > > Tehran, P.O. Box: 14395-547- IRAN > > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > > Cellphone: +989177317514 > > E-Mail: m.abbasnejad at gmail.com > > Website: http://physics.ut.ac.ir > > > > --------------------------------------------------------- > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > http://www.cmm.upenn.edu > > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA > 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a better > idiot. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > --------------------------------------------------------- > > Mohaddeseh Abbasnejad, > Room No. 323, Department of Physics, > University of Tehran, North Karegar Ave., > Tehran, P.O. Box: 14395-547- IRAN > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > Cellphone: +989177317514 > E-Mail: m.abbasnejad at gmail.com > Website: http://physics.ut.ac.ir > > --------------------------------------------------------- > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From akohlmey at cmm.chem.upenn.edu Tue May 12 22:24:13 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 12 May 2009 16:24:13 -0400 Subject: [Pw_forum] Hartree-Fock calculation In-Reply-To: <4A09D807.8070300@sissa.it> References: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> <1242150901.3135.35.camel@zero> <7699a1950905121308j3120e148w90ccf72fe5cd69d3@mail.gmail.com> <4A09D807.8070300@sissa.it> Message-ID: <1242159853.3135.64.camel@zero> On Tue, 2009-05-12 at 22:11 +0200, Stefano de Gironcoli wrote: > mohaddeseh abbasnejad wrote: > > Dear Axel, > > > > Honestly I heard it from one of my colleagues. In order to him, it's > > under test. He just knew about it and nothing more. > > indeed ... read example/EXX_example/README ciao stefano, well, i'd like to argue that a hybrid DFT calculation is not what people typically understand under a "hartree fock" calculation, i.e. something similar to what "traditional" quantum chemistry codes like gamess, gaussian, q-chem, and so on do using local basis sets. cheers, axel. > > Stefano de Gironcoli - SISSA and DEMOCRITOS > > > > Yours, > > Mohaddeseh > > > > > > > > On Tue, May 12, 2009 at 10:25 PM, Axel Kohlmeyer > > > wrote: > > > > On Tue, 2009-05-12 at 09:29 -0700, mohaddeseh abbasnejad wrote: > > > Dear all, > > > > > > I think that it's possible to do Hartree-Fock calculation using > > PWSCF > > > but it needs some utilities which should be installed before using. > > > > dear mohaddeseh, > > > > please explain where you get this impression from. > > > > > I was wondering if you could help me. > > > > no. > > > > > Thanks in advance. > > > > cheers, > > axel. > > > > > Yours, > > > Mohaddeseh > > > > > > > > > > > > > > > -- > > > --------------------------------------------------------- > > > > > > Mohaddeseh Abbasnejad, > > > Room No. 323, Department of Physics, > > > University of Tehran, North Karegar Ave., > > > Tehran, P.O. Box: 14395-547- IRAN > > > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > > > Cellphone: +989177317514 > > > E-Mail: m.abbasnejad at gmail.com > > > Website: http://physics.ut.ac.ir > > > > > > --------------------------------------------------------- > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > ======================================================================= > > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu > > http://www.cmm.upenn.edu > > > > Center for Molecular Modeling -- University of Pennsylvania > > Department of Chemistry, 231 S.34th Street, Philadelphia, PA > > 19104-6323 > > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > > ======================================================================= > > If you make something idiot-proof, the universe creates a better > > idiot. > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > -- > > --------------------------------------------------------- > > > > Mohaddeseh Abbasnejad, > > Room No. 323, Department of Physics, > > University of Tehran, North Karegar Ave., > > Tehran, P.O. Box: 14395-547- IRAN > > Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 > > Cellphone: +989177317514 > > E-Mail: m.abbasnejad at gmail.com > > Website: http://physics.ut.ac.ir > > > > --------------------------------------------------------- > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From rcjhawk at gmail.com Wed May 13 02:40:41 2009 From: rcjhawk at gmail.com (Michael Mehl) Date: Tue, 12 May 2009 20:40:41 -0400 Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <556671.31919.qm@web54403.mail.yahoo.com> References: <556671.31919.qm@web54403.mail.yahoo.com> Message-ID: <4A0A1709.1020307@gmail.com> Ah, but if you do that you'll have a left-handed coordinate system instead of a right-handed one. To convert from "unique axis c" coordinates to "unique axis b" coordinates, and simultaneously maintain the handedness of the primitive vectors, take a -> c b -> a c -> b alpha -> gamma beta -> alpha gamma -> beta HOWEVER: If you are in a base-centered monoclinic lattice, complications can ensue. In that case, study the various options in the space group tables, and be very, very, careful, making sure that you have the chosen the correct vectors to define the base in relation to the unique axis. A large number of the errors in the Lattice pages are because I wasn't careful about this. Jonas Baltrusaitis wrote: > I just went through an example of monoclinic crystal structure http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html > > Seems like when compared with description in INPUT_PW b should be swapped with c and beta should be swapped with gamma (going from cif file to pwscf input) > > Jonas > > -- Michael Mehl Naval Research Laboratory, Washington DC Author of http://cst-www.nrl.navy.mil/lattice (and other things) But I'm not at work now, so standard disclaimers apply From jasius_1 at yahoo.com Wed May 13 03:02:52 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 12 May 2009 18:02:52 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <4A0A1709.1020307@gmail.com> Message-ID: <930888.52436.qm@web54408.mail.yahoo.com> very useful. do I have to move around x,y,z coords as well? --- On Tue, 5/12/09, Michael Mehl wrote: > From: Michael Mehl > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 12, 2009, 5:40 PM > Ah, but if you do that you'll have a left-handed > coordinate system instead of a right-handed one. > > To convert from "unique axis c" coordinates to > "unique axis b" coordinates, and simultaneously > maintain the handedness of the primitive vectors, take > > a -> c > b -> a > c -> b > > alpha -> gamma > beta -> alpha > gamma -> beta > > HOWEVER: If you are in a base-centered monoclinic lattice, > complications can ensue. In that case, study the various > options in the space group tables, and be very, very, > careful, making sure that you have the chosen the correct > vectors to define the base in relation to the unique axis. > > A large number of the errors in the Lattice pages are > because I wasn't careful about this. > > > Jonas Baltrusaitis wrote: > > > I just went through an example of monoclinic crystal > structure > http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html > > > > Seems like when compared with description in INPUT_PW > b should be swapped with c and beta should be swapped with > gamma (going from cif file to pwscf input) > > > > Jonas > > > > > > > -- Michael Mehl > Naval Research Laboratory, Washington DC > Author of http://cst-www.nrl.navy.mil/lattice (and other > things) > But I'm not at work now, so standard disclaimers apply From jasius_1 at yahoo.com Wed May 13 03:08:15 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 12 May 2009 18:08:15 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <4A0A1709.1020307@gmail.com> Message-ID: <851520.80460.qm@web54410.mail.yahoo.com> wait, but then it's not clear anymore what angles are between which axes... Jonas --- On Tue, 5/12/09, Michael Mehl wrote: > From: Michael Mehl > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 12, 2009, 5:40 PM > Ah, but if you do that you'll have a left-handed > coordinate system instead of a right-handed one. > > To convert from "unique axis c" coordinates to > "unique axis b" coordinates, and simultaneously > maintain the handedness of the primitive vectors, take > > a -> c > b -> a > c -> b > > alpha -> gamma > beta -> alpha > gamma -> beta > > HOWEVER: If you are in a base-centered monoclinic lattice, > complications can ensue. In that case, study the various > options in the space group tables, and be very, very, > careful, making sure that you have the chosen the correct > vectors to define the base in relation to the unique axis. > > A large number of the errors in the Lattice pages are > because I wasn't careful about this. > > > Jonas Baltrusaitis wrote: > > > I just went through an example of monoclinic crystal > structure > http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html > > > > Seems like when compared with description in INPUT_PW > b should be swapped with c and beta should be swapped with > gamma (going from cif file to pwscf input) > > > > Jonas > > > > > > > -- Michael Mehl > Naval Research Laboratory, Washington DC > Author of http://cst-www.nrl.navy.mil/lattice (and other > things) > But I'm not at work now, so standard disclaimers apply From jasius_1 at yahoo.com Wed May 13 04:39:05 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Tue, 12 May 2009 19:39:05 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <4A0A1709.1020307@gmail.com> Message-ID: <187467.45970.qm@web54410.mail.yahoo.com> It just became increasingly confusing. Say I will rotate a,b,c and alpha, beta, gamma and x,y,z. But then pwscf "gamma", e.g. cosab gets rotated, too, so this 90 degree angle from cif again ends up as cosab again which it can't be Jonas --- On Tue, 5/12/09, Michael Mehl wrote: > From: Michael Mehl > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Tuesday, May 12, 2009, 5:40 PM > Ah, but if you do that you'll have a left-handed > coordinate system instead of a right-handed one. > > To convert from "unique axis c" coordinates to > "unique axis b" coordinates, and simultaneously > maintain the handedness of the primitive vectors, take > > a -> c > b -> a > c -> b > > alpha -> gamma > beta -> alpha > gamma -> beta > > HOWEVER: If you are in a base-centered monoclinic lattice, > complications can ensue. In that case, study the various > options in the space group tables, and be very, very, > careful, making sure that you have the chosen the correct > vectors to define the base in relation to the unique axis. > > A large number of the errors in the Lattice pages are > because I wasn't careful about this. > > > Jonas Baltrusaitis wrote: > > > I just went through an example of monoclinic crystal > structure > http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html > > > > Seems like when compared with description in INPUT_PW > b should be swapped with c and beta should be swapped with > gamma (going from cif file to pwscf input) > > > > Jonas > > > > > > > -- Michael Mehl > Naval Research Laboratory, Washington DC > Author of http://cst-www.nrl.navy.mil/lattice (and other > things) > But I'm not at work now, so standard disclaimers apply From ddrsmile at gmail.com Wed May 13 06:27:49 2009 From: ddrsmile at gmail.com (DdrSmiLe) Date: Wed, 13 May 2009 12:27:49 +0800 Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <245408.78234.qm@web65703.mail.ac4.yahoo.com> References: <245408.78234.qm@web65703.mail.ac4.yahoo.com> Message-ID: Dear Eyvaz Thank you for your reply It helps much. I noticed the code, lambda.f90, you shared contents the information about T_c. I am sorry for neglecting affiliation I belong to Sincerely, Joey, Liu Acadmic Sinica, IAMS Research Assistant 2009/5/13 Eyvaz Isaev > > Dear Liu, > > --- On Tue, 5/12/09, DdrSmiLe wrote: > > > Would your code print out the T_c? > > If so, where would it be printed? > > Without any doubts, if everything went properly. You can find T_c printed > after lambda.x finished. > > > My major is not chemistry or physics, but I'm doing job > > related to them. Therefore, there are too many things confuse me. > > > Then yo should read some solid state textbooks (Kittel, Ashcroft-Mermin, > Zaiman, Ricahrd Martin, Harrison, Kohanoff, Kantorovich, etc.) > This will help you not to be confused. > > Please provide your affiliation when post a question to this forum. > > Bests, > Eyvaz. > > > > Joey, Liu > > > > 2009/5/7 Eyvaz Isaev > > > > > Hi, > > > > > > Sorry for empty mail, unexpectedly I clicked on Send > > button. > > > > > > Well, I think attached files are what you need. The > > first one (modified > > > elphon.f90) writes special files, and the second one > > uses these files to > > > find lambda and then T_c via the Allen-Dynes equation > > (\omega_log instead of > > > the Debye temperature). > > > > > > In fact, recently I suggested to change original > > elphon.f90 and lambda.f90 > > > files with those (at least, include changes), but I do > > not know whether this > > > suggestion was accepted by developers. > > > Attached elphon.f90 worked for QE 4.0.4, at least. > > > > > > Any questions, please contact me. > > > > > > Bests, > > > Eyvaz. > > > > > > > > ------------------------------------------------------------------- > > > Prof. Eyvaz Isaev, > > > Theoretical Physics Department, Moscow State Institute > > of Steel & Alloys, > > > Russia, > > > Department of Physics, Chemistry, and Biology (IFM), > > Linkoping University, > > > Sweden > > > Condensed Matter Theory Group, Uppsala University, > > Sweden > > > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, > > eyvaz_isaev at yahoo.com > > > > > > > > > --- On Wed, 5/6/09, David Tompsett > > wrote: > > > > > > > From: David Tompsett > > > > Subject: [Pw_forum] Documentation on El-Ph > > coupling > > > > To: pw_forum at pwscf.org > > > > Date: Wednesday, May 6, 2009, 4:30 PM > > > > Dear All, > > > > I am interested in calculating the > > > > electron-phonon coupling > > > > of a metallic system in QE. I would like to then > > apply the > > > > McMillan > > > > formula for the prediction of the superconducting > > Tc. > > > > > > > > In the basic tutorials list, /Hands-on Tutorial > > of Quantum > > > > Espresso > > > > > > > > > > >,/ > > > > I could find > > > > examples on the El-ph calculation for Aluminium > > at X. Is > > > > there any > > > > further documentation available to indicated on > > how to find > > > > the > > > > integrated lambda and wlog that enter the > > McMillan formula > > > > for Tc? > > > > > > > > Thank you, > > > > David. > > > > > > > > -- > > > > David A. Tompsett > > > > Quantum Matter Group > > > > Cavendish Laboratory > > > > J. J. Thomson Avenue > > > > Cambridge CB3 0HE > > > > U.K. > > > > Tel: +44 7907 566351 (mobile) > > > > Fax: +44 1223 768140 > > > > http://www-qm.phy.cam.ac.uk/ > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/2197eee6/attachment-0001.htm From baroni at sissa.it Wed May 13 07:46:42 2009 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 13 May 2009 07:46:42 +0200 Subject: [Pw_forum] Hartree-Fock calculation In-Reply-To: <1242159853.3135.64.camel@zero> References: <7699a1950905120929h58a9bef3vd70d9193c2ac41cc@mail.gmail.com> <1242150901.3135.35.camel@zero> <7699a1950905121308j3120e148w90ccf72fe5cd69d3@mail.gmail.com> <4A09D807.8070300@sissa.it> <1242159853.3135.64.camel@zero> Message-ID: Hi All. In my understanding, "Hartree-Fock calculation" refers to the level of theory, and not to any specific basis set used to implement it. You can have "Hartree-Fock" with plane waves, "DFT" with gaussians, the first coded by a chemist, the latter by a physicist, or any sensible permutation. Hybrid-functional calculations, although strictly speking DFT, are technically very close to HF calculations, in the sense that any hybrid-functional code can easily produce HF results with whateve basis set it can use. All in all, PWscf can be used to perform HF calculations using plane waves as a basis set. I would agree arguing that PWs are probably not the most efficient basis set to implement non-locaal (Fock) exchange. Cheers - SB On May 12, 2009, at 10:24 PM, Axel Kohlmeyer wrote: > On Tue, 2009-05-12 at 22:11 +0200, Stefano de Gironcoli wrote: >> mohaddeseh abbasnejad wrote: >>> Dear Axel, >>> >>> Honestly I heard it from one of my colleagues. In order to him, it's >>> under test. He just knew about it and nothing more. >> >> indeed ... read example/EXX_example/README > > ciao stefano, > > well, i'd like to argue that a hybrid DFT calculation is not what > people typically understand under a "hartree fock" calculation, i.e. > something similar to what "traditional" quantum chemistry codes like > gamess, gaussian, q-chem, and so on do using local basis sets. > > > cheers, > axel. > > > >> >> Stefano de Gironcoli - SISSA and DEMOCRITOS >> >> >>> Yours, >>> Mohaddeseh >>> >>> >>> >>> On Tue, May 12, 2009 at 10:25 PM, Axel Kohlmeyer >>> > >>> wrote: >>> >>> On Tue, 2009-05-12 at 09:29 -0700, mohaddeseh abbasnejad wrote: >>>> Dear all, >>>> >>>> I think that it's possible to do Hartree-Fock calculation using >>> PWSCF >>>> but it needs some utilities which should be installed before using. >>> >>> dear mohaddeseh, >>> >>> please explain where you get this impression from. >>> >>>> I was wondering if you could help me. >>> >>> no. >>> >>>> Thanks in advance. >>> >>> cheers, >>> axel. >>> >>>> Yours, >>>> Mohaddeseh >>>> >>>> >>>> >>>> >>>> -- >>>> --------------------------------------------------------- >>>> >>>> Mohaddeseh Abbasnejad, >>>> Room No. 323, Department of Physics, >>>> University of Tehran, North Karegar Ave., >>>> Tehran, P.O. Box: 14395-547- IRAN >>>> Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 >>>> Cellphone: +989177317514 >>>> E-Mail: m.abbasnejad at gmail.com >>>> Website: http://physics.ut.ac.ir >>>> >>>> --------------------------------------------------------- >>>> >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >>> -- >>> >>> = >>> = >>> = >>> ==================================================================== >>> Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu >>> http://www.cmm.upenn.edu >>> >>> Center for Molecular Modeling -- University of Pennsylvania >>> Department of Chemistry, 231 S.34th Street, Philadelphia, PA >>> 19104-6323 >>> tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: >>> 1-215-898-5425 >>> >>> = >>> = >>> = >>> ==================================================================== >>> If you make something idiot-proof, the universe creates a better >>> idiot. >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >>> >>> >>> >>> -- >>> --------------------------------------------------------- >>> >>> Mohaddeseh Abbasnejad, >>> Room No. 323, Department of Physics, >>> University of Tehran, North Karegar Ave., >>> Tehran, P.O. Box: 14395-547- IRAN >>> Tel. No.: +98 21 6111 8634 & Fax No.: +98 21 8800 4781 >>> Cellphone: +989177317514 >>> E-Mail: m.abbasnejad at gmail.com >>> Website: http://physics.ut.ac.ir >>> >>> --------------------------------------------------------- >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > -- > = > ====================================================================== > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http:// > www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA > 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > = > ====================================================================== > If you make something idiot-proof, the universe creates a better > idiot. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/e18888e2/attachment-0001.htm From padmaja_patnaik at yahoo.co.uk Wed May 13 08:36:36 2009 From: padmaja_patnaik at yahoo.co.uk (Padmaja Patnaik) Date: Wed, 13 May 2009 06:36:36 +0000 (GMT) Subject: [Pw_forum] Doubt in magnetic behaviour Message-ID: <781712.98587.qm@web26103.mail.ukl.yahoo.com> Dear all i am calculating magnetic moment of SiC with transition metal as substitutional impurity considering supercells. I am doing with single atom substitution either at Si place or at C place. First i am doing the scf calculations then the nscf calculations to find out the density of states. Many published results shows that Cr with SiC shows ferromagnetic behaviour. But in my calculations i obtained magnetic momnet of the Cr atom as zero. The total magnetization of the sample is also giving zero value. Also the results obtained for Mn in SiC is quite different than previous published results. In Mn case i am getting very small magnetic moment which is quite less than the published values. Is anybody working on it?? Can anybody help me regarding this? Thanx in advance, Padmaja Patnaik Research Scholar Dept of Physics IIT Bombay Mumbai, India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/c6a7462d/attachment.htm From giuseppe.mattioli at mlib.ism.cnr.it Wed May 13 11:23:02 2009 From: giuseppe.mattioli at mlib.ism.cnr.it (Giuseppe Mattioli) Date: Wed, 13 May 2009 11:23:02 +0200 Subject: [Pw_forum] Doubt in magnetic behaviour In-Reply-To: <781712.98587.qm@web26103.mail.ukl.yahoo.com> References: <781712.98587.qm@web26103.mail.ukl.yahoo.com> Message-ID: <200905131123.03386.giuseppe.mattioli@mlib.ism.cnr.it> Dear Padmaja GGA-LSDA calculations show severe limits when treating transition metal impurities in a semiconductor host matrix. you should try some beyond-LDA approach like, e.g., the Hubbard U correction as implemented in QE. Yours Giuseppe On Wednesday 13 May 2009 08:36:36 Padmaja Patnaik wrote: > Dear all > > i am calculating magnetic moment of SiC with transition metal as > substitutional impurity considering supercells. I am doing with single atom > substitution either at Si place or at C place. First i am doing the scf > calculations then the nscf calculations to find out the density of states. > Many published results shows that Cr with SiC shows ferromagnetic > behaviour. But in my calculations i obtained magnetic momnet of the Cr atom > as zero. The total magnetization of the sample is also giving zero value. > Also the results obtained for Mn in SiC is quite different than previous > published results. In Mn case i am getting very small magnetic moment which > is quite less than the published values. Is anybody working on it? > Can anybody help me regarding this? > > Thanx in advance, > > Padmaja Patnaik > > Research Scholar > > Dept of Physics > > IIT Bombay > > Mumbai, India -- ******************************************************** - Article premier - Les hommes naissent et demeurent libres et ?gaux en droits. Les distinctions sociales ne peuvent ?tre fond?es que sur l'utilit? commune - Article 2 - Le but de toute association politique est la conservation des droits naturels et imprescriptibles de l'homme. Ces droits sont la libert?, la propri?t?, la s?ret? et la r?sistance ? l'oppression. ******************************************************** Giuseppe Mattioli CNR - ISTITUTO DI STRUTTURA DELLA MATERIA v. Salaria Km 29,300 - C.P. 10 I 00016 - Monterotondo Stazione (RM) Tel + 39 06 90672836 - Fax +39 06 90672316 E-mail: -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/00a90753/attachment.htm From paulatto at sissa.it Wed May 13 11:36:58 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 13 May 2009 11:36:58 +0200 (CEST) Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <851520.80460.qm@web54410.mail.yahoo.com> References: <851520.80460.qm@web54410.mail.yahoo.com> Message-ID: <50069.78.12.161.199.1242207418.squirrel@webmail.sissa.it> On Wed, May 13, 2009 03:08, Jonas Baltrusaitis wrote: > > wait, but then it's not clear anymore what angles are between which > axes... > > Jonas The "new" gamma is between the new a and the new b, while the "old" gamma is between the old a and the old b. It is just a redefinition, you are not changing your system your are just changing point of view. On the x,y,z swapping: it is a bit complicated to redefine the coordinates, at least I'm not sure it is a simple permutation. You've better define the atoms positions in "crystal" units, in this way you can just apply the same permutation you used to redefine the axis. But there may be a simpler solution: use ibrav=0 and define your axis by hand according to your preferred convention. It may sound dirty, but it is not: the code can still find all the symmetries operations. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From anvari_meh at physics.iust.ac.ir Wed May 13 15:59:05 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Wed, 13 May 2009 17:29:05 +0330 Subject: [Pw_forum] van der waals force Message-ID: Dear Paulatto I read your discussion about , how we can sure about stability of the one layer material. One of your suggestion was to compare the energy of isolated layer with a layer in bulk, I am working on graphene and in first step I relaxed my system and compared the result with graphite. the energy of graphite was higher than graphene. now my question is, there is van der waals bond in graphite and I've heard DFT calculations are not suitable for this materials because of its approximation in potential .Is it correct or some correction is implemented in Q-E. Also during my calculation I realized that when I increased the value of (celldm(3)=3-10) to attached isolated plane for graphene, there isn't any noticeable flactuations in total energy, but the values of subenergies such as hartree, ewald energy,...change extremely,I mean these terms become large and I think it is not logical.What do you think? Best Regards Mehrnaz Anvari Iran University Of Science & Technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/3c3402da/attachment.htm From jasius_1 at yahoo.com Wed May 13 15:08:30 2009 From: jasius_1 at yahoo.com (jasius_1 at yahoo.com) Date: Wed, 13 May 2009 06:08:30 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question Message-ID: <779084.83669.qm@web54402.mail.yahoo.com> thanks. the problem is that the new gamma between new a and b is still 90o! Right now in the cif file I have a unique angle between a and c, while angles between a and b and b and c are 90o. According to the pwscf manual, the only angle I need to specify is between a and b (that's why I started this rotation question). but if I rotate everything once then this angle becomes between rotated vectors b and c, still not a and b as requested in the manual. shall I rotate twice, e.g. 240 degrees? then the new vectors comprising the unique angle would actually be a and b, as requested in the manual. victory shall be mine! the problem with transofrming xyz is that I work with cartesians. I can get them easiliy and, to admit the truth, I still can't figure out the difference between alat and crystal coords. just can;t. read the Wyckoff file on the manual page, can't understand the difference is alat simple fractional coords? can they be simply rotated twice then? finally, what is a good check that all the rotations were valid? Jonas --- On Wed, 5/13/09, Lorenzo Paulatto wrote: > From: Lorenzo Paulatto > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Wednesday, May 13, 2009, 2:36 AM > > On Wed, May 13, 2009 03:08, Jonas Baltrusaitis wrote: > > > >? wait, but then it's not clear anymore what > angles are between which > >? axes... > > > >? Jonas > > > The "new" gamma is between the new a and the new b, while > the "old" gamma > is between the old a and the old b. It is just a > redefinition, you are not > changing your system your are just changing point of view. > > On the x,y,z swapping: it is a bit complicated to redefine > the > coordinates, at least I'm not sure it is a simple > permutation. You've > better define the atoms positions in "crystal" units, in > this way you can > just apply the same permutation you used to redefine the > axis. > > But there may be a simpler solution: use ibrav=0 and define > your axis by > hand according to your preferred convention. It may sound > dirty, but it is > not: the code can still find all the symmetries > operations. > > regards > > -- > Lorenzo Paulatto > SISSA? &? DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www:???http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > ? SISSA Webmail https://webmail.sissa.it/ > ? Powered by SquirrelMail http://www.squirrelmail.org/ > > From paulatto at sissa.it Wed May 13 16:33:40 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 13 May 2009 16:33:40 +0200 Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <779084.83669.qm@web54402.mail.yahoo.com> References: <779084.83669.qm@web54402.mail.yahoo.com> Message-ID: In data 13 maggio 2009 alle ore 15:08:30, ha scritto: > Right now in the cif file I have a unique angle between a and c, while > angles between a and b and b and c are 90o. According to the pwscf > manual, the only angle I need to specify is between a and b (that's why > I started this rotation question). but if I rotate everything once then > this angle becomes between rotated vectors b and c, still not a and b as > requested in the manual. I don't know enough about monoclinic systems to answer this question, furthermore I don't understand around what you are rotating! > the problem with transofrming xyz is that I work with cartesians. I can > get them easiliy and, to admit the truth, I still can't figure out the > difference between alat and crystal coords. Alat are x,y,z coordinates, using alat (celldm(0), or the length of a) as the unit of measure; on the other hand crystal expresses the position of the ions as a linear combination of the basis vectors. e.g. from the navy site where it says B1 = + x1 A1 + 1/4 A2 + z1 A3 it's using crystal coordinates, conversely B1 = [x1 a + z1 c cos(beta)] X + 1/4 b Y + z1 c sin(beta) Z are cartesian coordinates, they become alat coordinates if you substitute a=>1 and b=>b/a, c=>c/a. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From jasius_1 at yahoo.com Wed May 13 16:38:29 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Wed, 13 May 2009 07:38:29 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question Message-ID: <818108.34267.qm@web54410.mail.yahoo.com> Fair enough. Big question is: a unique angle ab in monoclinic system if transferred from cif file is 90o, e.g. not unique at all. one of the two remaining angles is unique. hence I am trying to rotate the cell so this unique angle comes into place, e.g. between a and b, as defined in pwscf manual. it the get's complicated if x,y,z, coords should get rotated as well Jonas --- On Wed, 5/13/09, Lorenzo Paulatto wrote: > From: Lorenzo Paulatto > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Date: Wednesday, May 13, 2009, 7:33 AM > In data 13 maggio 2009 alle ore > 15:08:30, > ha scritto: > > Right now in the cif file I have a unique angle > between a and c, while angles between a and b and b and c > are 90o. According to the pwscf manual, the only angle I > need to specify is between a and b (that's why I started > this rotation question). but if I rotate everything once > then this angle becomes between rotated vectors b and c, > still not a and b as requested in the manual. > > I don't know enough about monoclinic systems to answer this > question, furthermore I don't understand around what you are > rotating! > > > the problem with transofrming xyz is that I work with > cartesians. I can get them easiliy and, to admit the truth, > I still can't figure out the difference between alat and > crystal coords. > > Alat are x,y,z coordinates, using alat (celldm(0), or the > length of a) as the unit of measure; on the other hand > crystal expresses the position of the ions as a linear > combination of the basis vectors. > > e.g. from the navy site > where it says > B1?????= ??? + x1 > A1 + 1/4 A2 + z1 A3 > it's using crystal coordinates, conversely > B1??? =??? [x1 a + z1 c > cos(beta)] X + 1/4 b Y + z1 c sin(beta) Z > are cartesian coordinates, they become alat coordinates if > you substitute a=>1 and b=>b/a, c=>c/a. > > regards > > --Lorenzo Paulatto > SISSA? &? DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www:???http://people.sissa.it/~paulatto/ > > ? ? *** save italian brains *** > http://saveitalianbrains.wordpress.com/ > From paulatto at sissa.it Wed May 13 17:14:37 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Wed, 13 May 2009 17:14:37 +0200 Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <818108.34267.qm@web54410.mail.yahoo.com> References: <818108.34267.qm@web54410.mail.yahoo.com> Message-ID: In data 13 maggio 2009 alle ore 16:38:29, Jonas Baltrusaitis ha scritto: > > Fair enough. Big question is: a unique angle ab in monoclinic system if > transferred from cif file is 90o, e.g. not unique at all. one of the two > remaining angles is unique. hence I am trying to rotate the cell so this > unique angle comes into place, e.g. between a and b, as defined in pwscf > manual. it the get's complicated if x,y,z, coords should get rotated as > well Exactly, that's why I was suggesting you to express the coordinates in crystal units, so you can do a simple premutation instead of a rotation. What *I think* you can do, to solve the problem in a formal way is: 1. write the cell basis vectors as provided by the cif file (C) 2. write the cell basis vectors as used by pw.x (P) using the same units of measure 4. compute R=P.C^-1 5. apply R to all the atomic coordinates from the CIF (intended as column vectors) file to get the coordinates for pw.x Yet, I am more and more convinced that defining the cell axis by hand would be much simpler in this case. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From wshen02 at gmail.com Thu May 14 01:34:00 2009 From: wshen02 at gmail.com (=?ISO-2022-JP?B?GyRCSjhEQBsoQg==?=) Date: Wed, 13 May 2009 16:34:00 -0700 Subject: [Pw_forum] Dear all,Does anybody calculate Dysprosium? Message-ID: <2c9df8a80905131634s7241cda9uf389cf3636d1d5f8@mail.gmail.com> Dear All, Did anyone calculate Dysprosium before? Does anyone have the pseudopotential for Dysprosium? If you guys have the pseudopotential for Dysprosium, will you be so kind to send me a copy of that. Thanks a lot,guys! Wen Ph.D candidate Georgetown University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090513/401047a8/attachment.htm From rcjhawk at gmail.com Thu May 14 03:32:39 2009 From: rcjhawk at gmail.com (Michael Mehl) Date: Wed, 13 May 2009 21:32:39 -0400 Subject: [Pw_forum] monoclinic cell definition question In-Reply-To: <779084.83669.qm@web54402.mail.yahoo.com> References: <779084.83669.qm@web54402.mail.yahoo.com> Message-ID: <4A0B74B7.6090700@gmail.com> jasius_1 at yahoo.com wrote: > thanks. the problem is that the new gamma between new a and b is still 90o! Sigh. This is what comes of trying to answer a question near my bedtime. I'm afraid that I've totally fraked the explanation. Let's try again: The page http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html shows a lattice with unique axis "b", not "c", as is obvious from the "beta" angle. Last night I was thinking it was a "c" lattice, but like I said, it was near my bedtime. So the problem is: how to convert a monoclinic lattice from a unique axis b representation to a unique axis c representation, and visa versa, while preserving the shape of the unit cell. Apologies to the list, this is going to take a bit of time: By convention, when we say "a monoclinic lattice with unique axis b" we mean the second vector of the set of three primitive vectors is perpendicular to the other two, and those other two are not perpendicular, but are pointed at some non-90 degree angle relative to one another. So let's look at a set of lattice vectors for this system. One choice (and there are an infinite number) would be: a1 = [ s , 0 , 0 ] a2 = [ 0 , t , 0 ] a3 = [ u cos(v) , 0 , u sin(v) ] where v <> 90 degrees and I deliberately didn't use a,b,c,alpha,beta,gamma. Note that this is the choice made on the web page in question. Now in the traditional expression of lattice vectors: a = Sqrt[a1.a1] = s b = Sqrt[a2.a2] = t c = Sqrt[a3.a3] = u and the angles are defined by cos(alpha) = a2.a3/(|a2| |a3|) = 0 cos(beta) = a3.a1/(|a3| }a1|) = cos(v) => beta = v cos(gamma) = a1.a2/(|a1| |a2|) = 0 That's the unique axis b monoclinic lattice: vector b (a2) is perpendicular to vectors a (a1) and c (a3). Now let's take the same system and express it as a monoclinic system with unique axis c. That is, we want vector a3 to be perpendicular to a1 and a2, and a1 and a2 to have the same non-right angle between them as vectors a1 and a3 did in the old system, while preserving the shape of the primitive cell. One more or less obvious choice, which preserves the right-hand nature of the system, is to just permute (not rotate, duh) the vectors, and list them as: a1' = [ u cos(v) , 0 , u sin(v) ] a2' = [ s , 0 , 0 ] a3' = [ 0 , t , 0 ] And now: a = Sqrt[a1'.a1'] = u b = Sqrt[a2'.a2'] = s c = Sqrt[a3'.a3'] = t and cos(alpha) = a2'.a3'/(|a2'| |a3'|) = 0 cos(beta) = a3'.a1'/(|a3'| }a1'|) = 0 cos(gamma) = a1'.a2'/(|a1'| |a2'|) = cos(v) => gamma = v However (and this is where the "rotate" comment came in), my personal quirk is that (a1,a2,a3) should be as close as possible to (X,Y,Z). So NOW lets rotate that system about the (111) axis, taking Z->X->Y->Z. We get a1" = [ u sin(v) , u cos(v) , 0 ] a2" = [ 0 , s , 0 ] a3" = [ 0 , 0 , t ] where we leave it as an exercise for the reader to show that this is the same combination of (a,b,c,alpha,beta,gamma) as the primed vectors above. What about the basis of the lattice? Suppose we have a basis vector that was written in original crystal (aka lattice) coordinates as B1 = p a1 + q a2 + r a3 In Cartesian space, this becomes B1 = [p s + r u cos(v), q t , r sin(v) ] and a little bit of thought shows that this can be written in terms of the primed lattice as: B1 = r a1' + p a2' + q a3' which is the permutation that someone mentioned. If you then want to go to my rotated space, you get B1 = r a1" + p a2" + q a3" = [r u sin(v) , r u cos(v) + p s, q t] I hope this clears things up, but once again it's getting near my bedtime, so it's possible I've made another mistake. This is a lot easier to see if you draw some pictures of the individual sets of primitive vectors. You'll see that all the sets (a1,a2,a3), (a1',a2',a3') and (a1",a2",a3") define the same lattice, and the the corresponding B1 vector is at the same place in the primitive cell. -- Mike Mehl Naval Research Laboratory, Washington DC (At least until someone reads this post) Disclaimer: If you've been following this discussion, you should know to take everything I've said with a grain of salt. From ddrsmile at gmail.com Thu May 14 04:40:06 2009 From: ddrsmile at gmail.com (DdrSmiLe) Date: Thu, 14 May 2009 10:40:06 +0800 Subject: [Pw_forum] Documentation on El-Ph coupling In-Reply-To: <245408.78234.qm@web65703.mail.ac4.yahoo.com> References: <245408.78234.qm@web65703.mail.ac4.yahoo.com> Message-ID: Dear Eyvaz When I executed the lambda.x, it printed out the error messages. I enter the " ./lambda.x < elph_8k_example" The content of elph_8k_example was 10 0.12 1 8 0.0000000 0.0000000 0.0000000 1.00 -0.2500000 -0.2500000 0.2500000 8.00 -0.5000000 -0.5000000 0.5000000 4.00 0.0000000 0.0000000 0.5000000 6.00 -0.2500000 -0.2500000 0.7500000 24.00 -0.5000000 -0.5000000 1.0000000 12.00 0.0000000 0.0000000 1.0000000 3.00 -0.5000000 0.0000000 1.0000000 6.00 elph. 0.000000. 0.000000. 0.000000 elph.-0.250000. 0.250000.-0.250000 elph. 0.500000.-0.500000. 0.500000 elph. 0.000000. 0.500000. 0.000000 elph. 0.750000.-0.250000. 0.750000 elph. 0.500000. 0.000000. 0.500000 elph. 0.000000.-1.000000. 0.000000 elph.-0.500000.-1.000000. 0.000000 0.10 forrtl: severe (64): input conversion error, unit 4, file /var/tmp/ntliu/espresso-4.0.5-md/examples/example07/results/elph. 0.000000. 0.000000. 0.000000 I ran the example07 attached in quantum-espresso. I used intel fortran10.1.018 compiler and intel MKL10.0.4.023 to compile quantum espresso. Would you give me suggestion? Thank you Joey, Liu IAMS, Acadamic Sinica Research Assistant 2009/5/13 Eyvaz Isaev > > Dear Liu, > > --- On Tue, 5/12/09, DdrSmiLe wrote: > > > Would your code print out the T_c? > > If so, where would it be printed? > > Without any doubts, if everything went properly. You can find T_c printed > after lambda.x finished. > > > My major is not chemistry or physics, but I'm doing job > > related to them. Therefore, there are too many things confuse me. > > > Then yo should read some solid state textbooks (Kittel, Ashcroft-Mermin, > Zaiman, Ricahrd Martin, Harrison, Kohanoff, Kantorovich, etc.) > This will help you not to be confused. > > Please provide your affiliation when post a question to this forum. > > Bests, > Eyvaz. > > > > Joey, Liu > > > > 2009/5/7 Eyvaz Isaev > > > > > Hi, > > > > > > Sorry for empty mail, unexpectedly I clicked on Send > > button. > > > > > > Well, I think attached files are what you need. The > > first one (modified > > > elphon.f90) writes special files, and the second one > > uses these files to > > > find lambda and then T_c via the Allen-Dynes equation > > (\omega_log instead of > > > the Debye temperature). > > > > > > In fact, recently I suggested to change original > > elphon.f90 and lambda.f90 > > > files with those (at least, include changes), but I do > > not know whether this > > > suggestion was accepted by developers. > > > Attached elphon.f90 worked for QE 4.0.4, at least. > > > > > > Any questions, please contact me. > > > > > > Bests, > > > Eyvaz. > > > > > > > > ------------------------------------------------------------------- > > > Prof. Eyvaz Isaev, > > > Theoretical Physics Department, Moscow State Institute > > of Steel & Alloys, > > > Russia, > > > Department of Physics, Chemistry, and Biology (IFM), > > Linkoping University, > > > Sweden > > > Condensed Matter Theory Group, Uppsala University, > > Sweden > > > Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, > > eyvaz_isaev at yahoo.com > > > > > > > > > --- On Wed, 5/6/09, David Tompsett > > wrote: > > > > > > > From: David Tompsett > > > > Subject: [Pw_forum] Documentation on El-Ph > > coupling > > > > To: pw_forum at pwscf.org > > > > Date: Wednesday, May 6, 2009, 4:30 PM > > > > Dear All, > > > > I am interested in calculating the > > > > electron-phonon coupling > > > > of a metallic system in QE. I would like to then > > apply the > > > > McMillan > > > > formula for the prediction of the superconducting > > Tc. > > > > > > > > In the basic tutorials list, /Hands-on Tutorial > > of Quantum > > > > Espresso > > > > > > > > > > >,/ > > > > I could find > > > > examples on the El-ph calculation for Aluminium > > at X. Is > > > > there any > > > > further documentation available to indicated on > > how to find > > > > the > > > > integrated lambda and wlog that enter the > > McMillan formula > > > > for Tc? > > > > > > > > Thank you, > > > > David. > > > > > > > > -- > > > > David A. Tompsett > > > > Quantum Matter Group > > > > Cavendish Laboratory > > > > J. J. Thomson Avenue > > > > Cambridge CB3 0HE > > > > U.K. > > > > Tel: +44 7907 566351 (mobile) > > > > Fax: +44 1223 768140 > > > > http://www-qm.phy.cam.ac.uk/ > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090514/9a56d263/attachment.htm From jdai3 at mail.ustc.edu.cn Thu May 14 10:38:36 2009 From: jdai3 at mail.ustc.edu.cn (Jun Dai) Date: Thu, 14 May 2009 16:38:36 +0800 Subject: [Pw_forum] structure optimization with LDA+U Message-ID: <442290316.08634@ustc.edu.cn> Dear QE users, I tried to do a vc-relax optimization within LDA+U, but there's an error saying "pseudopotential not yet inserted", so is it possible to do optimizations with lda+u in QE, if yes, how? Thanks. ------------------------------ Best regards, Jun Dai ============================================================= J. Dai Ph.D. Candidate, Hefei National Laboratory For Physical Sciences at the Microscale, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 E-mail: jdai3 at mail.ustc.edu.cn ============================================================= From giuseppe.mattioli at mlib.ism.cnr.it Thu May 14 11:06:19 2009 From: giuseppe.mattioli at mlib.ism.cnr.it (Giuseppe Mattioli) Date: Thu, 14 May 2009 11:06:19 +0200 Subject: [Pw_forum] structure optimization with LDA+U In-Reply-To: <442290316.08634@ustc.edu.cn> References: <442290316.08634@ustc.edu.cn> Message-ID: <200905141106.20798.giuseppe.mattioli@mlib.ism.cnr.it> You must edit by hand the following files: YOUR_QE/PW/set_hubbard_l.f90 YOUR_QE/PW/tabd.f90 adding the lacking atoms and specifying the shell occupancies. Make some tests...! Yours Giuseppe On Thursday 14 May 2009 10:38:36 Jun Dai wrote: > Dear QE users, > I tried to do a vc-relax optimization within LDA+U, but there's an error > saying "pseudopotential not yet inserted", so is it possible to do > optimizations with lda+u in QE, if yes, how? Thanks. > > > > > ------------------------------ > Best regards, > Jun Dai > > ============================================================= > J. Dai > Ph.D. Candidate, > Hefei National Laboratory For Physical Sciences at the Microscale, > University of Science and Technology of China, > Hefei, Anhui, 230026, > People's Republic of China > Tel.: 86-551-3606428 > Fax.: 86-551-3602969 > E-mail: jdai3 at mail.ustc.edu.cn > ============================================================= > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ******************************************************** - Article premier - Les hommes naissent et demeurent libres et ?gaux en droits. Les distinctions sociales ne peuvent ?tre fond?es que sur l'utilit? commune - Article 2 - Le but de toute association politique est la conservation des droits naturels et imprescriptibles de l'homme. Ces droits sont la libert?, la propri?t?, la s?ret? et la r?sistance ? l'oppression. ******************************************************** Giuseppe Mattioli CNR - ISTITUTO DI STRUTTURA DELLA MATERIA v. Salaria Km 29,300 - C.P. 10 I 00016 - Monterotondo Stazione (RM) Tel + 39 06 90672836 - Fax +39 06 90672316 E-mail: -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090514/0879ea46/attachment.htm From pc229 at kent.ac.uk Thu May 14 11:08:30 2009 From: pc229 at kent.ac.uk (Pieremanuele Canepa) Date: Thu, 14 May 2009 10:08:30 +0100 Subject: [Pw_forum] run PWscf on national grid service in UK Message-ID: <8b443ba10905140208t6af6879h317800721d98eb3e@mail.gmail.com> Dear All, I am writing to you since I'm attempting to run PWscf jobs on the NGS (national grid service http://www.grid-support.ac.uk/) provide here in UK without any good result. Under the help of one person in charge of the ngs I was able to compile PWscf (release 4.0.5) on it. This grid use particular and non free mpi called Platform-mpi ( http://www.scali.com/Products/platform-mpi). Then, I compiled espresso using Intel FORTRAN and C compiler and these mpi libraries. Now when I try to run a job on the grid, what I get out is reported as below: Program PWSCF v.4.0.5 starts ... Today is 14May2009 at 9:23: 5 Parallel version (MPI) Number of processors in use: 8 R & G space division: proc/pool = 8 For Norm-Conserving or Ultrasoft (Vanderbilt) Pseudopotentials or PAW Current dimensions of program pwscf are: Max number of different atomic species (ntypx) = 10 Max number of k-points (npk) = 40000 Max angular momentum in pseudopotentials (lmaxx) = 3 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from read_namelists : error # 17 reading namelist control %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from read_namelists : error # 17 from read_namelists : error # 17 reading namelist control %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% TID HOST_NAME COMMAND_LINE STATUS TERMINATION_TIME ===== ========== ================ ======================= =================== 00000 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 00001 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:06 00002 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 00003 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 00004 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 00005 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 00006 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 00007 cn54.ngs.r /home/ngs0802/bi Exit (2) 05/14/2009 09:23:05 Initially, I thought it was related to my input and I tried to run it on my normal cluster PC, that use OpenMPI, and it seems to work as normally it should be. What I am suppose to do? Do you have any suggestion? Anyone do you know how to manage it ? Best Regards, Piero -- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom ----------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090514/fa7dc001/attachment.htm From paulatto at sissa.it Thu May 14 11:16:35 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 14 May 2009 11:16:35 +0200 Subject: [Pw_forum] run PWscf on national grid service in UK In-Reply-To: <8b443ba10905140208t6af6879h317800721d98eb3e@mail.gmail.com> References: <8b443ba10905140208t6af6879h317800721d98eb3e@mail.gmail.com> Message-ID: In data 14 maggio 2009 alle ore 11:08:30, Pieremanuele Canepa ha scritto: > Initially, I thought it was related to my input and I tried to run it > on my > normal cluster PC, that use OpenMPI, and it seems to work as normally it > should be. What I am suppose to do? Do you have any suggestion? Anyone do > you know how to manage it ? Dear Piero, for a start, you could check if you have feed the input to pw.x using stdin redirection "pw.x < input.file" or using the command line argument "pw.x -inp input.file". If you only tried the former, please try again with the latter syntax, which is generally safer. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From jdai3 at mail.ustc.edu.cn Thu May 14 11:20:13 2009 From: jdai3 at mail.ustc.edu.cn (Jun Dai) Date: Thu, 14 May 2009 17:20:13 +0800 Subject: [Pw_forum] structure optimization with LDA+U Message-ID: <442292813.09065@ustc.edu.cn> Thanks Giuseppe, I will have a try. ????????????????????: >From: Giuseppe Mattioli >Reply-To: PWSCF Forum >To: Jun Dai , PWSCF Forum >Subject: Re: [Pw_forum] structure optimization with LDA+U >Date:Thu, 14 May 2009 11:06:19 +0200 > > >You must edit by hand the following files: >YOUR_QE/PW/set_hubbard_l.f90 >YOUR_QE/PW/tabd.f90 >adding the lacking atoms and specifying the shell occupancies. >Make some tests...! > >Yours >Giuseppe > >On Thursday 14 May 2009 10:38:36 Jun Dai wrote: >> Dear QE users, >> I tried to do a vc-relax optimization within LDA+U, but there's an error >> saying "pseudopotential not yet inserted", so is it possible to do >> optimizations with lda+u in QE, if yes, how? Thanks. >> >> >> >> >> ------------------------------ >> Best regards, >> Jun Dai >> >> ============================================================= >> J. Dai >> Ph.D. Candidate, >> Hefei National Laboratory For Physical Sciences at the Microscale, >> University of Science and Technology of China, >> Hefei, Anhui, 230026, >> People's Republic of China >> Tel.: 86-551-3606428 >> Fax.: 86-551-3602969 >> E-mail: jdai3 at mail.ustc.edu.cn >> ============================================================= >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > >-- >******************************************************** >- Article premier - Les hommes naissent et demeurent >libres et ?gaux en droits. Les distinctions sociales >ne peuvent ?tre fond?es que sur l'utilit? commune >- Article 2 - Le but de toute association politique >est la conservation des droits naturels et >imprescriptibles de l'homme. Ces droits sont la libert?, >la propri?t?, la s?ret? et la r?sistance ? l'oppression. >******************************************************** > > Giuseppe Mattioli > CNR - ISTITUTO DI STRUTTURA DELLA MATERIA > v. Salaria Km 29,300 - C.P. 10 > I 00016 - Monterotondo Stazione (RM) > Tel + 39 06 90672836 - Fax +39 06 90672316 > E-mail: >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum ------------------------------ Best regards, Jun Dai ============================================================= J. Dai Ph.D. Candidate, Hefei National Laboratory For Physical Sciences at the Microscale, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 E-mail: jdai3 at mail.ustc.edu.cn ============================================================= From pc229 at kent.ac.uk Thu May 14 11:20:58 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Thu, 14 May 2009 10:20:58 +0100 Subject: [Pw_forum] run PWscf on national grid service in UK In-Reply-To: References: <8b443ba10905140208t6af6879h317800721d98eb3e@mail.gmail.com> Message-ID: Yes I did try? using the safe one without any? further result. Cheers ----- Original Message ----- From: Lorenzo?Paulatto? Date: Thursday, May 14, 2009 10:16 am Subject: Re: [Pw_forum] run PWscf?on national grid service in UK To: PWSCF?Forum > In data 14 maggio?2009 alle?ore 11:08:30, Pieremanuele? > Canepa? > ha scritto: > > Initially, I thought? it was related to my input and I > tried to run it? > > on my > > normal cluster PC, that use OpenMPI, and it seems to work as > normally it > > should be. What I am suppose to do? Do you have any > suggestion? Anyone do > > you know how to manage it ? > > > Dear Piero, > for a start, you could check if you have feed the input to pw.x > using? > stdin?redirection "pw.x < input.file" or using the command > line argument? > "pw.x -inp?input.file". If you only tried the former, please try > again? > with the latter syntax, which is generally safer. > > regards > > > -- > Lorenzo Paulatto > SISSA? &? DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www:?? http://people.sissa.it/~paulatto/ > > ???? *** save italian brains *** > ? http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090514/841cb64a/attachment.htm From paulatto at sissa.it Thu May 14 11:54:59 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 14 May 2009 11:54:59 +0200 Subject: [Pw_forum] run PWscf on national grid service in UK In-Reply-To: References: <8b443ba10905140208t6af6879h317800721d98eb3e@mail.gmail.com> Message-ID: In data 14 maggio 2009 alle ore 11:20:58, ha scritto: > Yes I did try? using the safe one without any? further result. > Cheers Ok, I just wanted to be sure. From what I see in the code, that may be the first place where a wrong call to mpi causes the code to stop, hence it may be that even if compilation went right the code is somewhere miscompiled or mislinked. It would be kind of you to try an replace the file Modules/read_namelist.f90 with the copy I'll send you in private (it is thousands of lines long) and recompile pw.x. I've only added a check to see if the read status is broadcasted correctly. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From pc229 at kent.ac.uk Thu May 14 11:56:44 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Thu, 14 May 2009 10:56:44 +0100 Subject: [Pw_forum] run PWscf on national grid service in UK In-Reply-To: References: <8b443ba10905140208t6af6879h317800721d98eb3e@mail.gmail.com> Message-ID: Thank you very much ----- Original Message ----- From: Lorenzo Paulatto Date: Thursday, May 14, 2009 10:55 am Subject: Re: [Pw_forum] run PWscf on national grid service in UK To: PWSCF Forum > In data 14 maggio 2009 alle ore 11:20:58, > ha scritto: > > Yes I did try? using the safe one without any? further result. > > Cheers > > Ok, I just wanted to be sure. > > ?From what I see in the code, that may be the first place > where a wrong? > call to mpi causes the code to stop, hence it may be that even > if? > compilation went right the code is somewhere miscompiled or > mislinked. It? > would be kind of you to try an replace the file > Modules/read_namelist.f90? > with the copy I'll send you in private (it is thousands of lines > long) and? > recompile pw.x. I've only added a check to see if the read > status is? > broadcasted correctly. > > regards > > > -- > Lorenzo Paulatto > SISSA? &? DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www:?? http://people.sissa.it/~paulatto/ > > ???? *** save italian brains *** > ? http://saveitalianbrains.wordpress.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090514/3e10d512/attachment.htm From sclauzer at sissa.it Thu May 14 12:25:59 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 14 May 2009 12:25:59 +0200 Subject: [Pw_forum] structure optimization with LDA+U In-Reply-To: <200905141106.20798.giuseppe.mattioli@mlib.ism.cnr.it> References: <442290316.08634@ustc.edu.cn> <200905141106.20798.giuseppe.mattioli@mlib.ism.cnr.it> Message-ID: <4A0BF1B7.1050608@sissa.it> Just a warning on top of this: if you're using UPF v2.x.x pseudopotentials and applying the U to atomic species with a single-character name (like "O" for instance), those subroutines may not recognize the species (even after it has been inserted). If so, either use UPF v1 or update to the latest CVS, where it should have been fixed. GS Giuseppe Mattioli wrote: > You must edit by hand the following files: > > YOUR_QE/PW/set_hubbard_l.f90 > > YOUR_QE/PW/tabd.f90 > > adding the lacking atoms and specifying the shell occupancies. > > Make some tests...! > > Yours > > Giuseppe > > On Thursday 14 May 2009 10:38:36 Jun Dai wrote: > > > Dear QE users, > > > I tried to do a vc-relax optimization within LDA+U, but there's an error > > > saying "pseudopotential not yet inserted", so is it possible to do > > > optimizations with lda+u in QE, if yes, how? Thanks. > > > > > > > > > > > > > > > ------------------------------ > > > Best regards, > > > Jun Dai > > > > > > ============================================================= > > > J. Dai > > > Ph.D. Candidate, > > > Hefei National Laboratory For Physical Sciences at the Microscale, > > > University of Science and Technology of China, > > > Hefei, Anhui, 230026, > > > People's Republic of China > > > Tel.: 86-551-3606428 > > > Fax.: 86-551-3602969 > > > E-mail: jdai3 at mail.ustc.edu.cn > > > ============================================================= > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > ******************************************************** > > - Article premier - Les hommes naissent et demeurent > > libres et ?gaux en droits. Les distinctions sociales > > ne peuvent ?tre fond?es que sur l'utilit? commune > > - Article 2 - Le but de toute association politique > > est la conservation des droits naturels et > > imprescriptibles de l'homme. Ces droits sont la libert?, > > la propri?t?, la s?ret? et la r?sistance ? l'oppression. > > ******************************************************** > > Giuseppe Mattioli > > CNR - ISTITUTO DI STRUTTURA DELLA MATERIA > > v. Salaria Km 29,300 - C.P. 10 > > I 00016 - Monterotondo Stazione (RM) > > Tel + 39 06 90672836 - Fax +39 06 90672316 > > E-mail: > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From jdai3 at mail.ustc.edu.cn Thu May 14 13:18:31 2009 From: jdai3 at mail.ustc.edu.cn (Jun Dai) Date: Thu, 14 May 2009 19:18:31 +0800 Subject: [Pw_forum] structure optimization with LDA+U Message-ID: <442299911.07175@ustc.edu.cn> Thanks Sclauzero, I will take care of this issue. ????????????????????: >From: Gabriele Sclauzero >Reply-To: PWSCF Forum >To: PWSCF Forum >Subject: Re: [Pw_forum] structure optimization with LDA+U >Date:Thu, 14 May 2009 12:25:59 +0200 > >Just a warning on top of this: if you're using UPF v2.x.x pseudopotentials and applying >the U to atomic species with a single-character name (like "O" for instance), those >subroutines may not recognize the species (even after it has been inserted). If so, either >use UPF v1 or update to the latest CVS, where it should have been fixed. > >GS > >Giuseppe Mattioli wrote: >> You must edit by hand the following files: >> >> YOUR_QE/PW/set_hubbard_l.f90 >> >> YOUR_QE/PW/tabd.f90 >> >> adding the lacking atoms and specifying the shell occupancies. >> >> Make some tests...! >> >> Yours >> >> Giuseppe >> >> On Thursday 14 May 2009 10:38:36 Jun Dai wrote: >> >> > Dear QE users, >> >> > I tried to do a vc-relax optimization within LDA+U, but there's an error >> >> > saying "pseudopotential not yet inserted", so is it possible to do >> >> > optimizations with lda+u in QE, if yes, how? Thanks. >> >> > >> >> > >> >> > >> >> > >> >> > ------------------------------ >> >> > Best regards, >> >> > Jun Dai >> >> > >> >> > ============================================================= >> >> > J. Dai >> >> > Ph.D. Candidate, >> >> > Hefei National Laboratory For Physical Sciences at the Microscale, >> >> > University of Science and Technology of China, >> >> > Hefei, Anhui, 230026, >> >> > People's Republic of China >> >> > Tel.: 86-551-3606428 >> >> > Fax.: 86-551-3602969 >> >> > E-mail: jdai3 at mail.ustc.edu.cn >> >> > ============================================================= >> >> > _______________________________________________ >> >> > Pw_forum mailing list >> >> > Pw_forum at pwscf.org >> >> > http://www.democritos.it/mailman/listinfo/pw_forum >> >> -- >> >> ******************************************************** >> >> - Article premier - Les hommes naissent et demeurent >> >> libres et ?gaux en droits. Les distinctions sociales >> >> ne peuvent ?tre fond?es que sur l'utilit? commune >> >> - Article 2 - Le but de toute association politique >> >> est la conservation des droits naturels et >> >> imprescriptibles de l'homme. Ces droits sont la libert?, >> >> la propri?t?, la s?ret? et la r?sistance ? l'oppression. >> >> ******************************************************** >> >> Giuseppe Mattioli >> >> CNR - ISTITUTO DI STRUTTURA DELLA MATERIA >> >> v. Salaria Km 29,300 - C.P. 10 >> >> I 00016 - Monterotondo Stazione (RM) >> >> Tel + 39 06 90672836 - Fax +39 06 90672316 >> >> E-mail: >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > >-- > > >o ------------------------------------------------ o >| Gabriele Sclauzero, PhD Student | >| c/o: SISSA & CNR-INFM Democritos, | >| via Beirut 2-4, 34014 Trieste (Italy) | >| email: sclauzer at sissa.it | >| phone: +39 040 3787 511 | >| skype: gurlonotturno | >o ------------------------------------------------ o >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum ------------------------------ Best regards, Jun Dai ============================================================= J. Dai Ph.D. Candidate, Hefei National Laboratory For Physical Sciences at the Microscale, University of Science and Technology of China, Hefei, Anhui, 230026, People's Republic of China Tel.: 86-551-3606428 Fax.: 86-551-3602969 E-mail: jdai3 at mail.ustc.edu.cn ============================================================= From kboryse at ncsu.edu Thu May 14 18:05:53 2009 From: kboryse at ncsu.edu (Kostyantyn Borysenko) Date: Thu, 14 May 2009 12:05:53 -0400 Subject: [Pw_forum] unitary matrix u Message-ID: <2C1C44865AA64068BE6465660B5DEA6E@ece.ncsu.edu> Hello all, I am trying to understand the meaning of some variables in the source code of QE. Particularly, it is not clear what is the meaning of the (3*nat, 3*nat) matrix u in phonon and electron-phonon calculations. In the output file generated by the subroutine phq_summary.f90 the columns of this matrix are referred to as "phonon polarizations" - i.e. they are supposed to be eigenvectors of the dynamical matrix. I consider the case when nat = 2 (two atoms in the unit cell), so the eigenvectors are 6-dimentional (3*nat=6) and there are six of them. I was told that in the case nat = 2 the components have the following meaning: (1x, 1y, 1z, 2x, 2y, 2z) - Cartesian components of deviations for both atoms. Now, in the output file generated by phq_summary.f90 for q=0, regardless of degeneracy of branches, there is always only one nonzero component. For instance, the first eigenvector is (-1,0,0,0,0,0). Obviously, all six eigenvectors are orthonormal, which is how it's supposed to be. But here is what bothers me. For acoustic phonon modes, the unit cell moves as a whole, i.e. both atoms must move with the same phase. So I expect to see something like (1/sqrt(2),0,0, 1/sqrt(2),0,0), instead of (-1,0,0,0,0,0). For optical branches on the other hand, two atoms have the opposite phase (the center of mass of the unit cell does not move), so I would expect to see something like (1/sqrt(2),0,0, -1/sqrt(2),0,0). My understanding is this: if every eigenvector has only one nonzero component, it means that in each mode, one atom is not moving at all! Maybe, I misunderstood the meaning of components of the matrix u? I apologize for this lengthy explanation. And I would really appreciate any help. Best regards, Kostyantyn Borysenko Electrical and Computer Engineering Department North Carolina State University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090514/27e23629/attachment-0001.htm From eyvaz_isaev at yahoo.com Fri May 15 01:18:17 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 14 May 2009 16:18:17 -0700 (PDT) Subject: [Pw_forum] Documentation on El-Ph coupling Message-ID: <501270.3702.qm@web65712.mail.ac4.yahoo.com> Dear Liu, It might be very helpful if you send me the whole info including input and output files. In order to keep clean the forum, please send them separately to my address only. Are these dyn* files in your current directory with lambda.x? Bests, Eyvaz. --- On Thu, 5/14/09, DdrSmiLe wrote: > From: DdrSmiLe > Subject: Re: [Pw_forum] Documentation on El-Ph coupling > To: eyvaz_isaev at yahoo.com, "PWSCF Forum" > Date: Thursday, May 14, 2009, 6:40 AM > Dear Eyvaz > ? > When I executed the lambda.x, it printed out the error > messages. > I enter the " ./lambda.x < > elph_8k_example" > ? > The content of elph_8k_example was > 10? 0.12? 1 > ??? 8 > ?????? 0.0000000? 0.0000000? 0.0000000?? 1.00 > ????? -0.2500000 -0.2500000? 0.2500000?? 8.00 > ????? -0.5000000 -0.5000000? 0.5000000?? 4.00 > ?????? 0.0000000? 0.0000000? 0.5000000?? 6.00 > > ????? -0.2500000 -0.2500000? 0.7500000? 24.00 > ????? -0.5000000 -0.5000000? 1.0000000? 12.00 > ?????? 0.0000000? 0.0000000? 1.0000000?? 3.00 > ????? -0.5000000? 0.0000000? 1.0000000?? 6.00 > elph. 0.000000. 0.000000. 0.000000 > > elph.-0.250000. 0.250000.-0.250000 > elph. 0.500000.-0.500000. 0.500000 > elph. 0.000000. 0.500000. 0.000000 > elph. 0.750000.-0.250000. 0.750000 > elph. 0.500000. 0.000000. 0.500000 > elph. 0.000000.-1.000000. 0.000000 > > elph.-0.500000.-1.000000. 0.000000 > 0.10 > > ? > forrtl: severe (64): input conversion error, unit 4, > file > /var/tmp/ntliu/espresso-4.0.5-md/examples/example07/results/elph. > 0.000000. 0.000000. 0.000000 > ? > I ran the example07 attached in > quantum-espresso. > I used intel fortran10.1.018 compiler and intel > MKL10.0.4.023?to compile quantum espresso. > > Would you give me suggestion? > ? > Thank you > > Joey, Liu > ? > IAMS, Acadamic Sinica > Research Assistant > ? > 2009/5/13 Eyvaz Isaev > > > Dear Liu, > > > --- On Tue, 5/12/09, DdrSmiLe > wrote: > > > Would your code print out the T_c? > > If so, where would it be printed? > > > Without any doubts, if everything went properly. You > can find T_c printed after lambda.x finished. > > > > My major is not chemistry or physics, but I'm > doing job > > related to them. Therefore, there are too many things > confuse me. > > > Then yo should read some solid state textbooks > (Kittel, Ashcroft-Mermin, Zaiman, Ricahrd Martin, Harrison, > Kohanoff, Kantorovich, etc.) > > This will help you not to be confused. > > Please provide your affiliation when post a question to > this forum. > > Bests, > Eyvaz. > > > > > > > Joey, Liu > > > > 2009/5/7 Eyvaz Isaev > > > > > Hi, > > > > > > Sorry for empty mail, unexpectedly I clicked on > Send > > > button. > > > > > > Well, I think attached files are what you need. > The > > first one (modified > > > elphon.f90) writes special files, and the second > one > > uses these files to > > > find lambda and then T_c via the Allen-Dynes > equation > > > (\omega_log instead of > > > the Debye temperature). > > > > > > In fact, recently I suggested to change original > > elphon.f90 and lambda.f90 > > > files with those (at least, include changes), but > I do > > > not know whether this > > > suggestion was accepted by developers. > > > Attached elphon.f90 worked for QE 4.0.4, at > least. > > > > > > Any questions, please contact me. > > > > > > Bests, > > > > Eyvaz. > > > > > > > > > ------------------------------------------------------------------- > > > Prof. Eyvaz Isaev, > > > Theoretical Physics Department, Moscow State > Institute > > > of Steel & Alloys, > > > Russia, > > > Department of Physics, Chemistry, and Biology > (IFM), > > Linkoping University, > > > Sweden > > > Condensed Matter Theory Group, Uppsala > University, > > > Sweden > > > Eyvaz.Isaev at fysik.uu.se, > isaev at ifm.liu.se, > > eyvaz_isaev at yahoo.com > > > > > > > > > > --- On Wed, 5/6/09, David Tompsett > > > wrote: > > > > > > > From: David Tompsett > > > > > Subject: [Pw_forum] Documentation on El-Ph > > coupling > > > > To: pw_forum at pwscf.org > > > > Date: Wednesday, May 6, 2009, 4:30 PM > > > ?> Dear All, > > > > > ? ? ? ? ? ? ?I am interested in > calculating the > > > > electron-phonon coupling > > > > of a metallic system in QE. I would like to > then > > apply the > > > > McMillan > > > > > formula for the prediction of the > superconducting > > Tc. > > > > > > > > In the basic tutorials list, /Hands-on > Tutorial > > of Quantum > > > > Espresso > > > > > > > > > > > > > > >,/ > > > > I could find > > > > examples on the El-ph calculation for > Aluminium > > at X. Is > > > > there any > > > > further documentation available to indicated > on > > > how to find > > > > the > > > > integrated lambda and wlog that enter the > > McMillan formula > > > > for Tc? > > > > > > > > Thank you, > > > > David. > > > > > > > > > -- > > > > David A. Tompsett > > > > Quantum Matter Group > > > > Cavendish Laboratory > > > > J. J. Thomson Avenue > > > > Cambridge CB3 0HE > > > > > U.K. > > > > Tel: +44 7907 566351 (mobile) > > > > Fax: +44 1223 768140 > > > > http://www-qm.phy.cam.ac.uk/ > > > > > > > > > > _______________________________________________ > > > > Pw_forum mailing list > > > > Pw_forum at pwscf.org > > > > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -----Inline Attachment Follows----- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From paulatto at sissa.it Fri May 15 11:13:51 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 15 May 2009 11:13:51 +0200 Subject: [Pw_forum] van der waals force In-Reply-To: References: Message-ID: In data 13 maggio 2009 alle ore 15:59:05, Mehrnaz Anvari ha scritto: > Is it correct or some correction is implemented in Q-E. Local density aproximation does not reproduce VdW bonds correctly, it can either overbound (like plain LDA usually does) or underbound (more common for gradien-corrected functionals). Yet, if you know the limits you can take your data with a grain of salt. At the moment QE does not implement any VdW correction by default, exact-exchange corrections are implemented but not compiled by default. > Also during my calculation I realized that when I increased the value of > (celldm(3)=3-10) to attached isolated plane for graphene, there isn't any > noticeable flactuations in total energy, but the values of subenergies > such as hartree, ewald energy,...change extremely,I mean these terms > become large and I think it is not logical. I don't see the problem, if you have a look a the definitions of the Ewald summation for the energy you can see it is perfectly normal. Of course the total energy has to converge for increasing size of the cell. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From physolde at yahoo.fr Fri May 15 12:14:42 2009 From: physolde at yahoo.fr (kada galami) Date: Fri, 15 May 2009 10:14:42 +0000 (GMT) Subject: [Pw_forum] elastic properties of orthorhombic structure Message-ID: <255390.91543.qm@web26605.mail.ukl.yahoo.com> dear all ?i want to know if we can calculate elastic properties of orthorhombic structure with pwscf code. thank all. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/531266f4/attachment.htm From eyvaz_isaev at yahoo.com Fri May 15 12:48:06 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 15 May 2009 03:48:06 -0700 (PDT) Subject: [Pw_forum] elastic properties of orthorhombic structure Message-ID: <608716.17271.qm@web65702.mail.ac4.yahoo.com> Dear Kalami, Of course, you can do that in QE (pscf), but applying strains and calculating total energies for these strained lattices. There is no direct way to calculate elastic constants in QE (I mean prepare an input file and submit a job). Please find the paper "First principles elastic constants and electronic structure of \alpha-Pt2Si and PtSi, O. Beckstein, J.E. Klepeis, G.L.W. Hart, and O.Pankratov" in Phys.Rev.B or cond-mat in arXiv.org (0008200v2) about strains to be applied. Please do not forget provide your affiliation. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Fri, 5/15/09, kada galami wrote: > From: kada galami > Subject: [Pw_forum] elastic properties of orthorhombic structure > To: Pw_forum at pwscf.org > Date: Friday, May 15, 2009, 2:14 PM > > dear all > > ?i want to know if we can calculate elastic properties > of orthorhombic structure with pwscf code. > > thank all. > > > > > > > > > -----Inline Attachment Follows----- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From yagufarov at phys.tsu.ru Fri May 15 13:24:33 2009 From: yagufarov at phys.tsu.ru (Ruslan S. Yagufarov) Date: Fri, 15 May 2009 18:24:33 +0700 Subject: [Pw_forum] Time of phonon calculation Message-ID: <20090515112404.M30045@phys.tsu.ru> Hello, dear Pwscf users, I have a question about phonon dispersion calculation in Ni2MnGa. The problem is that it takes too much time and the iteration procedure does not converge at all. I am tried to use various parameters, but without any success. After doing 100 iterations, Maybe I'm using wrong threshholds??? (my .scf and .ph inputs are attached below). As it seems to me? according to the previous post on this forum the everything is right. Actually, the questions are: 1)How much time does it usually take to calculate correct phonon dispersion for a system with only 4 atoms (like Ni2MnGa) in the case of non-parallel execution (assuming that you are using quite good computer with 2 processors and all parameters are good). Is it possible to have more than 50 hours calculation??? 2)Can anybody tell me what is wrong with this: SCF calculation &control calculation = 'scf', restart_mode='from_scratch', prefix='Ni2MnGa', pseudo_dir = '/home2/ruslan/q/pseudo', outdir= '/home2/ruslan/out' tprnfor=.true. tstress=.true. / &system ibrav= 2, celldm(1) =11.011 , nat = 4, ntyp = 3, ecutwfc = 27.0, ecutrho =350 , degauss=0.01 smearing = 'mp', occupations = 'smearing', nspin = 2, starting_magnetization(2) = 1.0 starting_magnetization(1) = 0.3 starting_magnetization(3) = 0 / &electrons / ATOMIC_SPECIES Ni 58.69 Ni.pbe-nd-rrkjus.UPF Mn 54.93 Mn.pbe-sp-van.UPF Ga 69.72 Ga.pbe-nsp-van.UPF ATOMIC_POSITIONS alat Ni 0.25 0.25 0.25 Ni -0.25 -0.25 -0.25 Mn -0.50 -0.50 -0.50 Ga 0.00 0.00 0.00 K_POINTS automatic 8 8 8 0 0 0 PH calculation &inputph prefix='Ni2MnGa', ldisp=.true., alpha_mix(1)=0.1, tr2_ph=1.0d-12, nq1=4, nq2=4, nq3=4, amass(1)=58.6943, amass(2)=54.938049, amass(3)=69.72, outdir='/home2/ruslan/out', fildyn='/home2/ruslan/out/ni2mnga.dynX', / Thank you! Best regards, Ruslan Tomsk State University, Institute of Material Science SB RAS From ptanhphys at gmail.com Fri May 15 14:09:36 2009 From: ptanhphys at gmail.com (TuanAnh Pham) Date: Fri, 15 May 2009 05:09:36 -0700 Subject: [Pw_forum] Time of phonon calculation In-Reply-To: <20090515112404.M30045@phys.tsu.ru> References: <20090515112404.M30045@phys.tsu.ru> Message-ID: Maybe because you use too many k-points, I guess. Try with a smaller number of k-points first, I've done a test with your input with 2 2 2 k-points and it's very fast. Hope this helps, On Fri, May 15, 2009 at 4:24 AM, Ruslan S. Yagufarov wrote: > Hello, dear Pwscf users, > > I have a question about phonon dispersion calculation in Ni2MnGa. > The problem is that it takes too much time and the iteration procedure does > not converge at all. I am tried to use various parameters, but without any > success. After doing 100 iterations, > Maybe I'm using wrong threshholds??? (my .scf and .ph inputs are attached > below). As it seems to me? according to the previous post on this forum the > everything is right. > > Actually, the questions are: > > 1)How much time does it usually take to calculate correct phonon dispersion > for a system with only 4 atoms (like Ni2MnGa) in the case of non-parallel > execution (assuming that you are using quite good computer with 2 > processors > and all parameters are good). Is it possible to have more than 50 hours > calculation??? > > 2)Can anybody tell me what is wrong with this: > > SCF calculation > > &control > calculation = 'scf', > restart_mode='from_scratch', > prefix='Ni2MnGa', > pseudo_dir = '/home2/ruslan/q/pseudo', > outdir= '/home2/ruslan/out' > tprnfor=.true. > tstress=.true. > / > &system > ibrav= 2, celldm(1) =11.011 , nat = 4, ntyp = 3, > ecutwfc = 27.0, ecutrho =350 , degauss=0.01 > smearing = 'mp', occupations = 'smearing', nspin = 2, > starting_magnetization(2) = 1.0 > starting_magnetization(1) = 0.3 > starting_magnetization(3) = 0 > / > &electrons > / > ATOMIC_SPECIES > Ni 58.69 Ni.pbe-nd-rrkjus.UPF > Mn 54.93 Mn.pbe-sp-van.UPF > Ga 69.72 Ga.pbe-nsp-van.UPF > ATOMIC_POSITIONS alat > Ni 0.25 0.25 0.25 > Ni -0.25 -0.25 -0.25 > Mn -0.50 -0.50 -0.50 > Ga 0.00 0.00 0.00 > K_POINTS automatic > 8 8 8 0 0 0 > > PH calculation > > &inputph > prefix='Ni2MnGa', > ldisp=.true., > alpha_mix(1)=0.1, > tr2_ph=1.0d-12, > nq1=4, > nq2=4, > nq3=4, > amass(1)=58.6943, > amass(2)=54.938049, > amass(3)=69.72, > outdir='/home2/ruslan/out', > fildyn='/home2/ruslan/out/ni2mnga.dynX', > / > > Thank you! > Best regards, Ruslan > Tomsk State University, Institute of Material Science SB RAS > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Tuan Anh Pham Graduate Student University of California, Davis Phone: 530-752-0957 Homepage: http://angstrom.ucdavis.edu/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/c177a97d/attachment.htm From panda.deng.pan at gmail.com Fri May 15 14:54:10 2009 From: panda.deng.pan at gmail.com (=?GB2312?B?xcu1xw==?=) Date: Fri, 15 May 2009 20:54:10 +0800 Subject: [Pw_forum] Time of phonon calculation In-Reply-To: References: <20090515112404.M30045@phys.tsu.ru> Message-ID: > >> alpha_mix(1)=0.1, >> >> This value seems too small.Maybe 0.3-0.5 would be better. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/d11defec/attachment.htm From marc at fq.edu.uy Fri May 15 14:36:30 2009 From: marc at fq.edu.uy (marc at fq.edu.uy) Date: Fri, 15 May 2009 09:36:30 -0300 (UYT) Subject: [Pw_forum] stack smashing error Message-ID: <51187.200.40.206.198.1242390990.squirrel@webmail.fq.edu.uy> Hello I have installed a quantum espresso 4.1 CVS copy on my computer to do some test, but when I try to run the program it fails. Configuration and compilation runs fine, but even examples gives this error. Previously I have installed 4.0.4 version which run smoothly, with examples. System is ubuntu 8.10. Many thanks Marc ------------------------------------------------------------- Marc E. Segovia Becker, PhD,Ch.Eng Computational Chemistry and Biology Group, Detema Facultad de Quimica Universidad de la Republica CC 1157A 11800 Montevideo Uruguay www.ccbg.fq.edu.uy -------------------------------------------------------------- Models are not reality. Nao se afobe nao, que nada ? pra ja. Chico If things seem under control, you're just not going fast enough.Andretti Soy jardinero de mis dilemas.Drexler *** stack smashing detected ***: ./bin/pw.x terminated ======= Backtrace: ========= > /lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb79b66d8] > /lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb79b6690] > ./bin/pw.x[0x823ce35] > ./bin/pw.x[0x80af77c] > ./bin/pw.x[0x80b6e90] > ./bin/pw.x[0x811379e] > ./bin/pw.x[0x81166b9] > ./bin/pw.x[0x804ab94] > ./bin/pw.x[0x83724c9] > /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb78d2685] > ./bin/pw.x[0x804aa71] > ======= Memory map: ======== > 08048000-08392000 r-xp 00000000 08:01 1182544 /home/marc/espresso/PW/pw.x > 08392000-08393000 r--p 00349000 08:01 1182544 /home/marc/espresso/PW/pw.x > 08393000-0839b000 rw-p 0034a000 08:01 1182544 /home/marc/espresso/PW/pw.x > 0839b000-09262000 rw-p 0839b000 00:00 0 > 0a92b000-0a94c000 rw-p 0a92b000 00:00 0 [heap] > b78a1000-b78a3000 rw-p b78a1000 00:00 0 > b78a3000-b78b8000 r-xp 00000000 08:02 1054 > /lib/tls/i686/cmov/libpthread-2.8.90.so > b78b8000-b78b9000 r--p 00014000 08:02 1054 > /lib/tls/i686/cmov/libpthread-2.8.90.so > b78b9000-b78ba000 rw-p 00015000 08:02 1054 > /lib/tls/i686/cmov/libpthread-2.8.90.so > b78ba000-b78bc000 rw-p b78ba000 00:00 0 > b78bc000-b7a14000 r-xp 00000000 08:02 1040 > /lib/tls/i686/cmov/libc-2.8.90.so > b7a14000-b7a16000 r--p 00158000 08:02 1040 > /lib/tls/i686/cmov/libc-2.8.90.so > b7a16000-b7a17000 rw-p 0015a000 08:02 1040 > /lib/tls/i686/cmov/libc-2.8.90.so > b7a17000-b7a1a000 rw-p b7a17000 00:00 0 > b7a1a000-b7a27000 r-xp 00000000 08:02 500972 /lib/libgcc_s.so.1 > b7a27000-b7a28000 r--p 0000c000 08:02 500972 /lib/libgcc_s.so.1 > b7a28000-b7a29000 rw-p 0000d000 08:02 500972 /lib/libgcc_s.so.1 > b7a29000-b7a4d000 r-xp 00000000 08:02 1044 > /lib/tls/i686/cmov/libm-2.8.90.so > b7a4d000-b7a4e000 r--p 00023000 08:02 1044 > /lib/tls/i686/cmov/libm-2.8.90.so > b7a4e000-b7a4f000 rw-p 00024000 08:02 1044 > /lib/tls/i686/cmov/libm-2.8.90.so > b7a4f000-b7b00000 r-xp 00000000 08:06 484923 > /usr/lib/libgfortran.so.3.0.0 > b7b00000-b7b01000 r--p 000b0000 08:06 484923 > /usr/lib/libgfortran.so.3.0.0 > b7b01000-b7b02000 rw-p 000b1000 08:06 484923 > /usr/lib/libgfortran.so.3.0.0 > b7b02000-b7b04000 rw-p b7b02000 00:00 0 > b7b04000-b7be4000 r-xp 00000000 08:06 488649 /usr/lib/libfftw3.so.3.1.2 > b7be4000-b7bea000 r--p 000df000 08:06 488649 /usr/lib/libfftw3.so.3.1.2 > b7bea000-b7beb000 rw-p 000e5000 08:06 488649 /usr/lib/libfftw3.so.3.1.2 > b7beb000-b7f62000 r-xp 00000000 08:06 18442 > /usr/lib/atlas/libblas.so.3gf.0 > b7f62000-b7f64000 r--p 00376000 08:06 18442 > /usr/lib/atlas/libblas.so.3gf.0 > b7f64000-b7f67000 rw-p 00378000 08:06 18442 > /usr/lib/atlas/libblas.so.3gf.0 > b7f7a000-b7f7c000 rw-p b7f7a000 00:00 0 > b7f7c000-b7f96000 r-xp 00000000 08:02 501022 /lib/ld-2.8.90.so > b7f96000-b7f97000 r-xp b7f96000 00:00 0 [vdso] > b7f97000-b7f98000 r--p 0001a000 08:02 501022 /lib/ld-2.8.90.so > b7f98000-b7f99000 rw-p 0001b000 08:02 501022 /lib/ld-2.8.90.so > bf984000-bf998000 rwxp bffeb000 00:00 0 [stack] > bf998000-bf999000 rw-p bffff000 00:00 0 > -- From padmaja_patnaik at yahoo.co.uk Fri May 15 15:32:08 2009 From: padmaja_patnaik at yahoo.co.uk (Padmaja Patnaik) Date: Fri, 15 May 2009 13:32:08 +0000 (GMT) Subject: [Pw_forum] Doubt in magnetic behaviour Message-ID: <104284.87902.qm@web26103.mail.ukl.yahoo.com> Thanks for your suggestion. I have another querry, while performing scf calculations for a semiconductor to which a transition metal has been added,? which occupations should be used, fixed or smearing? Thanks in advance Regards Padmaja Patnaik Research Scholar Dept of Physics IIT Bombay Mumbai, India -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/043f175e/attachment-0001.htm From akohlmey at cmm.chem.upenn.edu Fri May 15 15:45:44 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 15 May 2009 09:45:44 -0400 Subject: [Pw_forum] Doubt in magnetic behaviour In-Reply-To: <104284.87902.qm@web26103.mail.ukl.yahoo.com> References: <104284.87902.qm@web26103.mail.ukl.yahoo.com> Message-ID: <1242395144.2969.3.camel@zero> On Fri, 2009-05-15 at 13:32 +0000, Padmaja Patnaik wrote: dear padmaja, this is a good question to try and answer for yourself. you will be rewarded with a much better understanding of what you are doing and why. just consider what smearing is used for in the first place and whether, in principle, you can principle get the same (quality) result with fixed occupations as well. good luck, axel. > > > I have another querry, while performing scf calculations for a > semiconductor to which a transition metal has been added, which > occupations should be used, fixed or smearing? > > Thanks in advance > Regards > > > Padmaja Patnaik > Research Scholar > Dept of Physics > IIT Bombay > Mumbai, India > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Fri May 15 15:51:09 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 15 May 2009 09:51:09 -0400 Subject: [Pw_forum] stack smashing error In-Reply-To: <51187.200.40.206.198.1242390990.squirrel@webmail.fq.edu.uy> References: <51187.200.40.206.198.1242390990.squirrel@webmail.fq.edu.uy> Message-ID: <1242395469.2969.9.camel@zero> On Fri, 2009-05-15 at 09:36 -0300, marc at fq.edu.uy wrote: > Hello > hello marc, > I have installed a quantum espresso 4.1 CVS copy on my computer to do > some test, but when I try to run the program it fails. Configuration and > compilation runs fine, but even examples gives this error. that can mean two things: a) there is a problem with the compiler miscompiling new code that was added between since the 4.0.x version was branched off b) there is a bug in the code. both are equally probably. the cvs code is development code, so it tends to break every once in a while. and newly written code tends to use more fortran 95 features, which in turn can put a higher strain on compilers, particularly gfortran. i suggest you try compiling with reduced optimization and debugging symbols included and see if the error report persists and then produce a proper stack trace to help trying to pinpoint the offending part of the code, so that somebody else can have a look at it and determine who is to blame. cheers, axel. > Previously I have installed 4.0.4 version which run smoothly, with examples. > System is ubuntu 8.10. > Many thanks > > Marc > ------------------------------------------------------------- > Marc E. Segovia Becker, PhD,Ch.Eng > Computational Chemistry and Biology Group, Detema > Facultad de Quimica > Universidad de la Republica > CC 1157A > 11800 Montevideo > Uruguay > www.ccbg.fq.edu.uy > -------------------------------------------------------------- > > Models are not reality. > > Nao se afobe nao, que nada ? pra ja. Chico > > If things seem under control, you're just not going fast enough.Andretti > > Soy jardinero de mis dilemas.Drexler > > > *** stack smashing detected ***: ./bin/pw.x terminated > ======= Backtrace: ========= > > /lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb79b66d8] > > /lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb79b6690] > > ./bin/pw.x[0x823ce35] > > ./bin/pw.x[0x80af77c] > > ./bin/pw.x[0x80b6e90] > > ./bin/pw.x[0x811379e] > > ./bin/pw.x[0x81166b9] > > ./bin/pw.x[0x804ab94] > > ./bin/pw.x[0x83724c9] > > /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb78d2685] > > ./bin/pw.x[0x804aa71] > > ======= Memory map: ======== > > 08048000-08392000 r-xp 00000000 08:01 1182544 > /home/marc/espresso/PW/pw.x > > 08392000-08393000 r--p 00349000 08:01 1182544 > /home/marc/espresso/PW/pw.x > > 08393000-0839b000 rw-p 0034a000 08:01 1182544 > /home/marc/espresso/PW/pw.x > > 0839b000-09262000 rw-p 0839b000 00:00 0 > > 0a92b000-0a94c000 rw-p 0a92b000 00:00 0 [heap] > > b78a1000-b78a3000 rw-p b78a1000 00:00 0 > > b78a3000-b78b8000 r-xp 00000000 08:02 1054 > > /lib/tls/i686/cmov/libpthread-2.8.90.so > > b78b8000-b78b9000 r--p 00014000 08:02 1054 > > /lib/tls/i686/cmov/libpthread-2.8.90.so > > b78b9000-b78ba000 rw-p 00015000 08:02 1054 > > /lib/tls/i686/cmov/libpthread-2.8.90.so > > b78ba000-b78bc000 rw-p b78ba000 00:00 0 > > b78bc000-b7a14000 r-xp 00000000 08:02 1040 > > /lib/tls/i686/cmov/libc-2.8.90.so > > b7a14000-b7a16000 r--p 00158000 08:02 1040 > > /lib/tls/i686/cmov/libc-2.8.90.so > > b7a16000-b7a17000 rw-p 0015a000 08:02 1040 > > /lib/tls/i686/cmov/libc-2.8.90.so > > b7a17000-b7a1a000 rw-p b7a17000 00:00 0 > > b7a1a000-b7a27000 r-xp 00000000 08:02 500972 /lib/libgcc_s.so.1 > > b7a27000-b7a28000 r--p 0000c000 08:02 500972 /lib/libgcc_s.so.1 > > b7a28000-b7a29000 rw-p 0000d000 08:02 500972 /lib/libgcc_s.so.1 > > b7a29000-b7a4d000 r-xp 00000000 08:02 1044 > > /lib/tls/i686/cmov/libm-2.8.90.so > > b7a4d000-b7a4e000 r--p 00023000 08:02 1044 > > /lib/tls/i686/cmov/libm-2.8.90.so > > b7a4e000-b7a4f000 rw-p 00024000 08:02 1044 > > /lib/tls/i686/cmov/libm-2.8.90.so > > b7a4f000-b7b00000 r-xp 00000000 08:06 484923 > > /usr/lib/libgfortran.so.3.0.0 > > b7b00000-b7b01000 r--p 000b0000 08:06 484923 > > /usr/lib/libgfortran.so.3.0.0 > > b7b01000-b7b02000 rw-p 000b1000 08:06 484923 > > /usr/lib/libgfortran.so.3.0.0 > > b7b02000-b7b04000 rw-p b7b02000 00:00 0 > > b7b04000-b7be4000 r-xp 00000000 08:06 488649 /usr/lib/libfftw3.so.3.1.2 > > b7be4000-b7bea000 r--p 000df000 08:06 488649 /usr/lib/libfftw3.so.3.1.2 > > b7bea000-b7beb000 rw-p 000e5000 08:06 488649 /usr/lib/libfftw3.so.3.1.2 > > b7beb000-b7f62000 r-xp 00000000 08:06 18442 > > /usr/lib/atlas/libblas.so.3gf.0 > > b7f62000-b7f64000 r--p 00376000 08:06 18442 > > /usr/lib/atlas/libblas.so.3gf.0 > > b7f64000-b7f67000 rw-p 00378000 08:06 18442 > > /usr/lib/atlas/libblas.so.3gf.0 > > b7f7a000-b7f7c000 rw-p b7f7a000 00:00 0 > > b7f7c000-b7f96000 r-xp 00000000 08:02 501022 /lib/ld-2.8.90.so > > b7f96000-b7f97000 r-xp b7f96000 00:00 0 [vdso] > > b7f97000-b7f98000 r--p 0001a000 08:02 501022 /lib/ld-2.8.90.so > > b7f98000-b7f99000 rw-p 0001b000 08:02 501022 /lib/ld-2.8.90.so > > bf984000-bf998000 rwxp bffeb000 00:00 0 [stack] > > bf998000-bf999000 rw-p bffff000 00:00 0 > > > -- > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From marc at fq.edu.uy Fri May 15 15:58:22 2009 From: marc at fq.edu.uy (marc at fq.edu.uy) Date: Fri, 15 May 2009 10:58:22 -0300 (UYT) Subject: [Pw_forum] stack smashing error In-Reply-To: <1242395469.2969.9.camel@zero> References: <51187.200.40.206.198.1242390990.squirrel@webmail.fq.edu.uy> <1242395469.2969.9.camel@zero> Message-ID: <45819.200.40.206.198.1242395902.squirrel@webmail.fq.edu.uy> Hi Axel Many thanks for your advice, regards.marc > On Fri, 2009-05-15 at 09:36 -0300, marc at fq.edu.uy wrote:> Hello> > hello marc, >> I have installed a quantum espresso 4.1 CVS copy on my computer to do> >> some test, but when I try to run the program it fails. Configuration >> and> compilation runs fine, but even examples gives this error. > that can mean two things: > a) there is a problem with the compiler miscompiling new code that was > added between since the 4.0.x version was branched off > b) there is a bug in the code. > both are equally probably. the cvs code is development code, so ittends to > break every once in a while. and newly written code tends touse more > fortran 95 features, which in turn can put a higher strainon compilers, > particularly gfortran. > i suggest you try compiling with reduced optimization and debuggingsymbols > included and see if the error report persists and then producea proper > stack trace to help trying to pinpoint the offending part ofthe code, so > that somebody else can have a look at it and determinewho is to blame. > cheers, axel. > >> Previously I have installed 4.0.4 version which run smoothly, with >> examples.> System is ubuntu 8.10.> Many thanks> > Marc> >> -------------------------------------------------------------> Marc E. >> Segovia Becker, PhD,Ch.Eng> Computational Chemistry and Biology Group, >> Detema> Facultad de Quimica> Universidad de la Republica> CC 1157A> >> 11800 Montevideo> Uruguay> www.ccbg.fq.edu.uy> >> --------------------------------------------------------------> > >> Models are not reality.> > Nao se afobe nao, que nada ?? pra ja. >> Chico> > If things seem under control, you're just not going fast >> enough.Andretti> > Soy jardinero de mis dilemas.Drexler> > > *** >> stack smashing detected ***: ./bin/pw.x terminated> ======= Backtrace: >> =========> > >> /lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb79b66d8]> > >> /lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb79b6690]> > >> ./bin/pw.x[0x823ce35]> > ./bin/pw.x[0x80af77c]> > >> ./bin/pw.x[0x80b6e90]> > ./bin/pw.x[0x811379e]> > >> ./bin/pw.x[0x81166b9]> > ./bin/pw.x[0x804ab94]> > >> ./bin/pw.x[0x83724c9]> > >> /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb78d2685]> > >> ./bin/pw.x[0x804aa71]> > ======= Memory map: ========> > >> 08048000-08392000 r-xp 00000000 08:01 1182544 > >> /home/marc/espresso/PW/pw.x> > 08392000-08393000 r--p 00349000 08:01 >> 1182544 > /home/marc/espresso/PW/pw.x> > 08393000-0839b000 rw-p >> 0034a000 08:01 1182544 > /home/marc/espresso/PW/pw.x> > >> 0839b000-09262000 rw-p 0839b000 00:00 0> > 0a92b000-0a94c000 rw-p >> 0a92b000 00:00 0 [heap]> > b78a1000-b78a3000 rw-p b78a1000 >> 00:00 0> > b78a3000-b78b8000 r-xp 00000000 08:02 1054> > >> /lib/tls/i686/cmov/libpthread-2.8.90.so> > b78b8000-b78b9000 r--p >> 00014000 08:02 1054> > /lib/tls/i686/cmov/libpthread-2.8.90.so> > >> b78b9000-b78ba000 rw-p 00015000 08:02 1054> > >> /lib/tls/i686/cmov/libpthread-2.8.90.so> > b78ba000-b78bc000 rw-p >> b78ba000 00:00 0> > b78bc000-b7a14000 r-xp 00000000 08:02 1040> > >> /lib/tls/i686/cmov/libc-2.8.90.so> > b7a14000-b7a16000 r--p 00158000 >> 08:02 1040> > /lib/tls/i686/cmov/libc-2.8.90.so> > b7a16000-b7a17000 >> rw-p 0015a000 08:02 1040> > /lib/tls/i686/cmov/libc-2.8.90.so> > >> b7a17000-b7a1a000 rw-p b7a17000 00:00 0> > b7a1a000-b7a27000 r-xp >> 00000000 08:02 500972 /lib/libgcc_s.so.1> > b7a27000-b7a28000 r--p >> 0000c000 08:02 500972 /lib/libgcc_s.so.1> > b7a28000-b7a29000 rw-p >> 0000d000 08:02 500972 /lib/libgcc_s.so.1> > b7a29000-b7a4d000 r-xp >> 00000000 08:02 1044> > /lib/tls/i686/cmov/libm-2.8.90.so> > >> b7a4d000-b7a4e000 r--p 00023000 08:02 1044> > >> /lib/tls/i686/cmov/libm-2.8.90.so> > b7a4e000-b7a4f000 rw-p 00024000 >> 08:02 1044> > /lib/tls/i686/cmov/libm-2.8.90.so> > b7a4f000-b7b00000 >> r-xp 00000000 08:06 484923> > /usr/lib/libgfortran.so.3.0.0> > >> b7b00000-b7b01000 r--p 000b0000 08:06 484923> > >> /usr/lib/libgfortran.so.3.0.0> > b7b01000-b7b02000 rw-p 000b1000 08:06 >> 484923> > /usr/lib/libgfortran.so.3.0.0> > b7b02000-b7b04000 rw-p >> b7b02000 00:00 0> > b7b04000-b7be4000 r-xp 00000000 08:06 488649 >> /usr/lib/libfftw3.so.3.1.2> > b7be4000-b7bea000 r--p 000df000 08:06 >> 488649 /usr/lib/libfftw3.so.3.1.2> > b7bea000-b7beb000 rw-p >> 000e5000 08:06 488649 /usr/lib/libfftw3.so.3.1.2> > >> b7beb000-b7f62000 r-xp 00000000 08:06 18442> > >> /usr/lib/atlas/libblas.so.3gf.0> > b7f62000-b7f64000 r--p 00376000 >> 08:06 18442> > /usr/lib/atlas/libblas.so.3gf.0> > b7f64000-b7f67000 >> rw-p 00378000 08:06 18442> > /usr/lib/atlas/libblas.so.3gf.0> > >> b7f7a000-b7f7c000 rw-p b7f7a000 00:00 0> > b7f7c000-b7f96000 r-xp >> 00000000 08:02 501022 /lib/ld-2.8.90.so> > b7f96000-b7f97000 r-xp >> b7f96000 00:00 0 [vdso]> > b7f97000-b7f98000 r--p 0001a000 >> 08:02 501022 /lib/ld-2.8.90.so> > b7f98000-b7f99000 rw-p 0001b000 >> 08:02 501022 /lib/ld-2.8.90.so> > bf984000-bf998000 rwxp bffeb000 >> 00:00 0 [stack]> > bf998000-bf999000 rw-p bffff000 00:00 0> >> >> --> > > > > _______________________________________________> Pw_forum >> mailing list> Pw_forum at pwscf.org> >> http://www.democritos.it/mailman/listinfo/pw_forum > -- > =======================================================================Axel > Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of PennsylvaniaDepartment > of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323tel: > 1-215-898-1582, fax: 1-215-573-6233, office-tel: > 1-215-898-5425=======================================================================If > you make something idiot-proof, the universe creates a better idiot. > _______________________________________________Pw_forum mailing > listPw_forum at pwscf.orghttp://www.democritos.it/mailman/listinfo/pw_forum > From giuseppe.mattioli at mlib.ism.cnr.it Fri May 15 16:23:43 2009 From: giuseppe.mattioli at mlib.ism.cnr.it (Giuseppe Mattioli) Date: Fri, 15 May 2009 16:23:43 +0200 Subject: [Pw_forum] Doubt in magnetic behaviour In-Reply-To: <104284.87902.qm@web26103.mail.ukl.yahoo.com> References: <104284.87902.qm@web26103.mail.ukl.yahoo.com> Message-ID: <200905151623.44513.giuseppe.mattioli@mlib.ism.cnr.it> > While performing scf calculations for a > semiconductor to which a transition metal has been added,? which > occupations should be used, fixed or smearing? Your question doesn't have an easy answer. I'll try to share my (little and quite empirical) knowledge. If you add some gaussian smearing (a value of 0.01 Ry or less) your calculations will converge faster, but your electron density could be "delocalized" amongst Kohn-Sham wavefunction, the eigenvalues of which are placed across the Fermi energy of your system. This is particularly true when smearing value is too large wrt the energy difference between occupied and unoccupied bands, e.g., when your dopant atom induce rather shallow levels in the semiconductor band structure. For example, you could find that an occupied dopant level is "pinned" to the conduction band because of the smearing contribution, i.e., its K-S eigenvalue position and the total energy of your system depends upon the smearing value; On the other hand, such systems hardly converge if you use fixed occupations. At every electronic step your system starts to dance between, e.g., a configuration in which one electron is located on the impurity and another configuration in which the electron is located on the conduction band minimum. This is very painful when you have a molecule placed above (i.e., non interacting with) a semiconductor surface... So, as usual, there is no free lunch... Make several tests. I can only say that I always try to use smearing occupations, because of the convergence issue. But look always at the smearing contribution to the total energy. If it affects heavily the total energy value, then don't bet too much money on your results...:-) Further comments and suggestions are welcome... Yours Giuseppe On Friday 15 May 2009 15:32:08 Padmaja Patnaik wrote: > Thanks for your suggestion. > > I have another querry, while performing scf calculations for a > semiconductor to which a transition metal has been added,? which > occupations should be used, fixed or smearing? > > Thanks in advance > Regards > > > Padmaja Patnaik > > Research Scholar > > Dept of Physics > > IIT Bombay > > Mumbai, India -- ******************************************************** - Article premier - Les hommes naissent et demeurent libres et ?gaux en droits. Les distinctions sociales ne peuvent ?tre fond?es que sur l'utilit? commune - Article 2 - Le but de toute association politique est la conservation des droits naturels et imprescriptibles de l'homme. Ces droits sont la libert?, la propri?t?, la s?ret? et la r?sistance ? l'oppression. ******************************************************** Giuseppe Mattioli CNR - ISTITUTO DI STRUTTURA DELLA MATERIA v. Salaria Km 29,300 - C.P. 10 I 00016 - Monterotondo Stazione (RM) Tel + 39 06 90672836 - Fax +39 06 90672316 E-mail: -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/d81d9eff/attachment-0001.htm From yccheng.nju at gmail.com Fri May 15 16:29:20 2009 From: yccheng.nju at gmail.com (=?UTF-8?B?56iL6L+O5pil?=) Date: Fri, 15 May 2009 22:29:20 +0800 Subject: [Pw_forum] Time of phonon calculation In-Reply-To: <20090515112404.M30045@phys.tsu.ru> References: <20090515112404.M30045@phys.tsu.ru> Message-ID: > &electrons > / In the scf calculation, the default convergence threshold for scf conv_thr is 1.D-6, but in the phonon calculation, the threshold > tr2_ph=1.0d-12, > is much tighter !!! I think you can increase the scf threshold and decrease the phonon threshold. You can change the both threshold to 1.D-8. -- Y. C. Cheng Department of Physics Nanjing University Nanjing 210093 P. R. China Tel: 86-25-83592907 Email: yccheng.nju at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/2d0a2b19/attachment.htm From sclauzer at sissa.it Fri May 15 17:09:55 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 15 May 2009 17:09:55 +0200 Subject: [Pw_forum] Time of phonon calculation In-Reply-To: References: <20090515112404.M30045@phys.tsu.ru> Message-ID: <741FFEBD-C8EE-4C5B-8E7B-3094B91BD0CD@sissa.it> Il giorno 15/mag/09, alle ore 16:29, ??? ha scritto: > > > > &electrons > / > In the scf calculation, the default convergence threshold for scf > conv_thr is 1.D-6, but in the phonon calculation, the threshold > tr2_ph=1.0d-12, > is much tighter !!! Actually the converge thresholds for pw.x (conv_thr) and ph.x (tr2_ph) are not referred to the same quantity. The first is about the estimated error on the total energy, while the second is a sort of norm of the difference between the scf potential at two consecutive steps of the scf cycle (though I'm not completely sure). > > I think you can increase the scf threshold and decrease the phonon > threshold. You can change the both threshold to 1.D-8. In order to get good phonons it is usually better to have a well converged scf potential, so that it is good advice to decrease conv_thr down to 1.d-8. However, to my knowledge, it is better not to increase tr2_ph with respect to the default (1.d-12). GS > > > -- > Y. C. Cheng > Department of Physics > Nanjing University > Nanjing 210093 > P. R. China > Tel: 86-25-83592907 > Email: yccheng.nju at gmail.com > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum Gabriele Sclauzero, PhD Student SISSA & CNR-INFM Democritos, via Beirut 2-4, 34014 Trieste (Italy) email: sclauzer at sissa.it phone: +39 040 3787 511 skype: gurlonotturno -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/c9473dc7/attachment.htm From eyvaz_isaev at yahoo.com Fri May 15 17:10:35 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 15 May 2009 08:10:35 -0700 (PDT) Subject: [Pw_forum] elastic properties of orthorhombic structure Message-ID: <532181.9395.qm@web65709.mail.ac4.yahoo.com> Dear Galami, > Dear Kalami, Ooops, I have misprinted your name. Sorry. Bests, Eyvaz. --- On Fri, 5/15/09, Eyvaz Isaev wrote: > From: Eyvaz Isaev > Subject: Re: [Pw_forum] elastic properties of orthorhombic structure > To: "PWSCF Forum" > Date: Friday, May 15, 2009, 2:48 PM > > Dear Kalami, > > Of course, you can do that in QE (pscf), but applying > strains and calculating total energies for these strained > lattices. > There is no direct way to calculate elastic constants in QE > (I mean prepare an input file and submit a job). > Please find the paper "First principles elastic constants > and electronic structure of \alpha-Pt2Si and PtSi, O. > Beckstein, J.E. Klepeis, G.L.W. Hart, and O.Pankratov" in > Phys.Rev.B or cond-mat in arXiv.org (0008200v2) about > strains to be applied. > > Please do not forget provide your affiliation. > > Bests, > Eyvaz. > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of > Steel & Alloys, Russia, > Department of Physics, Chemistry, and Biology (IFM), > Linkoping University, Sweden > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, > isaev at ifm.liu.se, > eyvaz_isaev at yahoo.com > > > --- On Fri, 5/15/09, kada galami > wrote: > > > From: kada galami > > Subject: [Pw_forum] elastic properties of orthorhombic > structure > > To: Pw_forum at pwscf.org > > Date: Friday, May 15, 2009, 2:14 PM > > > > dear all > > > > ?i want to know if we can calculate elastic > properties > > of orthorhombic structure with pwscf code. > > > > thank all. > > > > > > > > > > > > > > > >? ? ??? > > -----Inline Attachment Follows----- > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > ? ? ? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From baroni at sissa.it Fri May 15 17:09:44 2009 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 15 May 2009 17:09:44 +0200 Subject: [Pw_forum] elastic properties of orthorhombic structure In-Reply-To: <255390.91543.qm@web26605.mail.ukl.yahoo.com> References: <255390.91543.qm@web26605.mail.ukl.yahoo.com> Message-ID: <471C5B67-E3AC-48E6-85DE-C44D7E1C7859@sissa.it> sure, you can please, sign your correspondence with this mailing list cheers - SB On May 15, 2009, at 12:14 PM, kada galami wrote: > > dear all > > i want to know if we can calculate elastic properties of > orthorhombic structure with pwscf code. > > thank all. > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090515/bd3267a9/attachment.htm From panda.deng.pan at gmail.com Sat May 16 10:58:42 2009 From: panda.deng.pan at gmail.com (=?GB2312?B?xcu1xw==?=) Date: Sat, 16 May 2009 16:58:42 +0800 Subject: [Pw_forum] question about elphonon calculation Message-ID: Dear All Users, I was trying to calculate the el-ph coupling.And I followed just as the example 07 did. First I run a dense scf calculation and then a scf calculation. Sencond I run the phonon calculation and it got error message. I attached the input of mine and the error message below.Although from the error message I thought there should be a phonon calculation before the elph calculation.But it run the elph calculation directely after the scf run have been done in example07 . I wonder that should I need a phonon calculation before the el-ph calculation after the scf run. And if the answer is yes, why it do not run in example07. Have a nice day Deng Pan Nanjing University elph.in MgB2-elph &inputph amass(1)= 24.3, amass(2)= 10.8, prefix='MgB2' tr2_ph=1.0d-14, alpha_mix(1)=0.5, iverbosity=1, fildyn='/MgB2/elph/test/MgB2.dyn', outdir='/tmp1/', ldisp=.true, nq1=4,nq2=4,nq3=4, elph=.true./ fildvscf='/MgB2/elph/test/MgB2dv' / and the error message task # 3 from phq_readin : error # 1 El-ph needs a DeltaVscf file -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090516/46f36b1b/attachment.htm From anvari_meh at physics.iust.ac.ir Sat May 16 14:27:54 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Sat, 16 May 2009 15:57:54 +0330 Subject: [Pw_forum] van der waals force Message-ID: Dear paulatto Thanks a lot because of your attention. I want to put graphene on substrate so I think that the van der waals force becomes important for this case (isn't correct?). By the way I have read the paper of M.Lazzeri & et al(Phys. Rev. B 78, 081406(R) (2008)) & I know the problem of LDA & GGA approximations(for graphene & graphite), so if there is any way to correct them please aware me. Well, I have to explain more about my first question. please assume that we have two layers of graphene. First the distance between them is 3 a.u & then it has been changed to 10 a.u. What will happen for hartree(Uee) term in calculations, Do you expect that it will increase or decrease? Best Regards Mehrnaz Anvari Iran University Of Science & Technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090516/bba612da/attachment.htm From akohlmey at cmm.chem.upenn.edu Sat May 16 14:08:50 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 16 May 2009 08:08:50 -0400 Subject: [Pw_forum] van der waals force In-Reply-To: References: Message-ID: <1242475730.2918.29.camel@zero> On Sat, 2009-05-16 at 15:57 +0330, Mehrnaz Anvari wrote: > Dear paulatto > Thanks a lot because of your attention. I want to put graphene on > substrate so I think that the van der waals force becomes important > for this case (isn't correct?). By the way I have read the paper of > M.Lazzeri & et al(Phys. Rev. B 78, 081406(R) (2008)) & I know the > problem of LDA & GGA approximations(for graphene & graphite), so if > there is any way to correct them please aware me. mehrnaz, several approaches have been tried to improve the representation of dispersion interactions in DFT. they range from hybrid-DFT, over additional correction terms derived from the self-consistent density to empirically modified pseudopotentials and adding classical force field terms. they vary in computational effort and packages where they have been implemented in. which of the methods are "best" depends a lot on what you would like to compute and how much computational power you have available. however, one should note that the methods with more computational effort do not always improve the results over "cheap" ones in certain cases. one has to understand how a given correction tries to address the problem to be able to make a choice. in short: there is a lot of literature out there. dig in! cheers, axel. > Well, I have to explain more about my first question. please assume > that we have two layers of graphene. First the distance between them > is 3 a.u & then it has been changed to 10 a.u. What will happen for > hartree(Uee) term in calculations, Do you expect that it will increase > or decrease? > Best Regards > Mehrnaz Anvari > Iran University Of Science & Technology > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From wangweiphysics at yahoo.com.cn Sat May 16 18:37:29 2009 From: wangweiphysics at yahoo.com.cn (wangwei) Date: Sun, 17 May 2009 00:37:29 +0800 (CST) Subject: [Pw_forum] The scf can't converge Message-ID: <451229.61290.qm@web15207.mail.cnb.yahoo.com> Dear everyone: ???????? Recently , I am calculating a material ,but It can't reach my convergence standard.I have tried many times ,but failed. My input file is : ? ???????????????????? ecutwfc = 50, ???????????????????? ecutrho = 500, ???????????????? occupations = 'smearing', ???????????????????? degauss = 0.03 ??????????????????? smearing = 'methfessel-paxton' , ?/ ?&ELECTRONS ??????????????????? conv_thr = 1.0D-8 , ???????????????? mixing_beta = 0.05 ,Please show me what's the matter ,Thank you ! Wei Wang?, Southeast University of NanJing City ___________________________________________________________ ????????????????? http://card.mail.cn.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090517/4ec58d0a/attachment.htm From paulatto at sissa.it Sat May 16 18:46:25 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 16 May 2009 18:46:25 +0200 (CEST) Subject: [Pw_forum] The scf can't converge In-Reply-To: <451229.61290.qm@web15207.mail.cnb.yahoo.com> References: <451229.61290.qm@web15207.mail.cnb.yahoo.com> Message-ID: <49623.78.12.169.199.1242492385.squirrel@webmail.sissa.it> Dear Wei Wang, > degauss = 0.03 this is a bit high, if you can afford more kpoints I would suggest to use a lower value. > mixing_beta = 0.05 this is extremely low, with such a low mixing beta there is no chance the code converges in less than 20 iteration! I cannot provide further detail without information on the system you are trying to compute. Sorry. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From andrea.vittadini at unipd.it Sun May 17 09:13:42 2009 From: andrea.vittadini at unipd.it (Andrea Vittadini) Date: Sun, 17 May 2009 09:13:42 +0200 Subject: [Pw_forum] van der waals force (Mehrnaz Anvarir) In-Reply-To: References: Message-ID: <4A0FB926.4030603@unipd.it> Mehrnaz, look for the post by Daniel Forrer (about two weeks ago, if I remember well). Regards, Andrea --------------------------------------------------- Istituto di Scienze e Tecnologie Molecolari del CNR Dip. Scienze Chimiche Universita' di Padova, Italy. email: andrea.vittadini at unipd.it From greatnaol at gmail.com Sun May 17 12:01:54 2009 From: greatnaol at gmail.com (Naol Regassa) Date: Sun, 17 May 2009 13:01:54 +0300 Subject: [Pw_forum] Hubbard U and small degauss ... Message-ID: <97e4b88e0905170301n7f8356b3x9b14c76ed44e4dd2@mail.gmail.com> Dear all I have two questions to ask a) For the last times I was calculating the effect of Hubbard U on the system YMnO_3 using QESPRESSO 4.10. I got an insulator band structure using U = 10.5eV though it is completely metal without U (GGA). When I got this result I have used small degauss = 0.006. My friend told me to use degauss = 0.015. Please tell me how possible (correct) is that? I checked the convergence for degauss but it is increasing near 0.015. b) Is adding U value more effective for the most stable structure ? For the unstable states like for my case cubic ferromagnetic phase I can't see any change? Here is the input for the 'scf' &control calculation = 'scf', restart_mode = 'from_scratch', prefix = 'ymo_hcp_ferro' tstress = .true., tprnfor = .true., pseudo_dir = '/home/MSC608/Desktop/QESPRESSO_PP', outdir='/home/MSC608/Thesis/Hexagonal/Ferro-Hex/temp' / &system ibrav = 0, celldm(1) = 6.7, nat = 10, ntyp = 4, ecutwfc = 32, ecutrho = 340.0 nspin = 2 starting_magnetization(2) = 0.7, starting_magnetization(3) = 0.7, occupations = 'smearing', smearing = 'marzari-vanderbilt', degauss = 0.005, lda_plus_u=.true., Hubbard_U(2) = 10.5, Hubbard_U(3) = 10.5, / &electrons diagonalization = 'david' startingwfc = 'atomic+random' mixing_beta = 0.3, conv_thr = 1.0d-10 / ATOMIC_SPECIES Y 88.91 Y.pbe-nsp-van.UPF Mn1 54.94 Mn.pbe-sp-van_mit.UPF Mn2 54.94 Mn.pbe-sp-van_mit.UPF O 15.9994 O.pbe-rrkjus.UPF CELL_PARAMETERS (alat) 0.985236441 -0.266531554 -0.000316189 -0.266366323 0.985859979 0.000349871 0.000781869 -0.000890980 3.186836665 ATOMIC_POSITIONS (crystal) Y -0.082663260 -0.082649328 0.000174846 Y -0.082874577 -0.082880779 0.499813069 Mn1 0.205305857 0.560245090 0.249986248 Mn2 0.561457263 0.207812583 0.750012603 O -0.027464907 -0.056329933 0.249979343 O -0.058482388 -0.031289340 0.750035308 O 0.247951998 0.582250730 0.083800805 O 0.582338704 0.248188482 0.916046842 O 0.582135578 0.247948396 0.583964864 O 0.247742343 0.582041831 0.416186070 K_POINTS automatic 3 3 2 1 1 1 Thanks alot in advance Tamene Regassa AAU Ethiopia From akohlmey at cmm.chem.upenn.edu Sun May 17 15:09:19 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 17 May 2009 09:09:19 -0400 Subject: [Pw_forum] The scf can't converge In-Reply-To: <451229.61290.qm@web15207.mail.cnb.yahoo.com> References: <451229.61290.qm@web15207.mail.cnb.yahoo.com> Message-ID: <1242565759.10549.20.camel@zero> On Sun, 2009-05-17 at 00:37 +0800, wangwei wrote: dear wangwei, ...and others have explained many, many times, that with only a few pieces of an input file, it is impossible to give any good advice. most likely either the problem is in the parts that you don't show, or you trying something that cannot work. when you go to a bakery you don't just say, "give me some of the brown stuff", do you? cheers, axel. > Dear everyone: > > Recently , I am calculating a material ,but It can't reach my > convergence standard.I have tried many times ,but failed. > My input file is : > > ecutwfc = 50, > ecutrho = 500, > occupations = 'smearing', > degauss = 0.03 > smearing = 'methfessel-paxton' , > / > &ELECTRONS > conv_thr = 1.0D-8 , > mixing_beta = 0.05 ,Please show me what's the > matter ,Thank you ! > Wei Wang , Southeast University of NanJing City > > > > ______________________________________________________________________ > ????????????????? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From wangweiphysics at yahoo.com.cn Sun May 17 17:18:03 2009 From: wangweiphysics at yahoo.com.cn (wangweiphysics at yahoo.com.cn) Date: Sun, 17 May 2009 23:18:03 +0800 (CST) Subject: [Pw_forum] The scf can't converge Message-ID: <149263.61872.qm@web15203.mail.cnb.yahoo.com> Dear Paulatto, Thanks? a lot for your reply! ? > mixing_beta = 0.05 >this is extremely low, with such a low mixing beta there is no chance the >code converges in less than 20 iteration! ?In fact, I set mixing_beta = 0.7 (defualt), the scf can't converge within 100 cycles. The troubleshooting in the userguide said that reduce mixing_beta from the default value to 0.3-0.1 or smaller, so I set mixing_beta = 0.2, and 0.05, both are also failed. > degauss = 0.03 >this is a bit high, if you can afford more kpoints I would suggest to use >a lower value. deduce it to 0.01, also failed. ? I have another queation: I don't know the difference between atomic position in units of alat and in crystal coordinates. When I use the crystal coordinates, scf can converge easily, but the atom positions in the output file are different from?those in the?input file. The same atom positins?obtain using atomic position in units of alat but can't converge.I am very puzzled! ? Wang wei, Southeast University, Nanjing, China 2009.5.17 ? ___________________________________________________________ ????????????????? http://card.mail.cn.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090517/a7f26ba5/attachment.htm From akohlmey at cmm.chem.upenn.edu Sun May 17 17:51:04 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 17 May 2009 11:51:04 -0400 Subject: [Pw_forum] The scf can't converge In-Reply-To: <149263.61872.qm@web15203.mail.cnb.yahoo.com> References: <149263.61872.qm@web15203.mail.cnb.yahoo.com> Message-ID: <1242575464.10549.212.camel@zero> On Sun, 2009-05-17 at 23:18 +0800, wangweiphysics at yahoo.com.cn wrote: [...] dear wang wei, > I have another queation: > I don't know the difference between atomic position in units of alat > and in crystal coordinates. When I use the crystal coordinates, scf > can converge easily, but the atom positions in the output file are > different from those in the input file. The same atom positins obtain i doubt that. please provide an example. have you compared input and output coordinates in xcrysden? > using atomic position in units of alat but can't converge.I am very > puzzled! you see, this is _exactly_ the reason, why you should post your complete input in the first place. bad initial atom positions is one of the possible reasons for problems with SCF convergence, as not physically meaningful atom positions will make a convergence very difficult. the difference of alat coordinates and crystal is very simple: alat are cartesian coordinates in units of the lattice constant, whereas crystal are fractional positions within the unitcell. for a primitive cubic lattice, they are identical, but i guess this is not what you are using. cheers, axel. > Wang wei, > Southeast University, Nanjing, China > 2009.5.17 > > > ______________________________________________________________________ > ????????????????? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From mozaharphysics25 at yahoo.com Sun May 17 17:49:17 2009 From: mozaharphysics25 at yahoo.com (Mozahar Ali) Date: Sun, 17 May 2009 08:49:17 -0700 (PDT) Subject: [Pw_forum] question about elphonon calculation Message-ID: <635360.31053.qm@web58507.mail.re3.yahoo.com> Dear all userI feel problem to calculate electron-phonon interaction with q-points.How can I choose the q-point mesh?The elph.output result is giving the error messegs as q- is not a vector in the dense grid.My scf,scf.fit and elph.in input files are given below and the last page of elph.out output also given.At first I run the scf.in and then scf.fit.in and finally elph.in.My sample is a superhard material.scf input:?&control ? prefix='BC5' ? restart_mode='from_scratch', ? calculation = 'scf' ? pseudo_dir = '/Thesis2007/espresso-3.1.1/pseudo' ? outdir='/Thesis2007/temp' ? tprnfor=.true. ? tstress = .true. ?/ ?&system ? ? ibrav= 4, celldm(1)=4.825803501, celldm(3)=2.5056, nat=6, ntyp= 2, ? ecutwfc =35.0, ecutrho=150.0, ? occupations='smearing', smearing='m-v',degauss =0.01, ?/ ?&electrons ? diagonalization= 'cg' , ? mixing_mode = 'plain' ? mixing_beta = 0.7 ? conv_thr = 1.0d-8 ?/ ATOMIC_SPECIES B 10.81000 B.pw91-n-van_ak.UPF? C 12.0107 C.pw91-van_ak.UPF ATOMIC_POSITIONS {crystal} ? B 0.000000000 0.000000000 0.007000000 ? ? C 0.000000000 0.000000000 0.259200000 ? ? C 0.333333300 -0.333333300 0.338600000 ? C 0.333333300 -0.333333300 0.581600000 ? C 0.666666667 -0.666666667 0.664600000 ? C 0.666666667 -0.666666667 0.899900000 ? ?K_POINTS automatic? ?9 9 4 0 0 0 scf.fit input: ?&control ? calculation ='scf' ? restart_mode='from_scratch', ? prefix='BC5' ? pseudo_dir = '/Thesis2007/espresso-3.1.1/pseudo' ? outdir='/Thesis2007/temp' ?/ ?&system ? ? ibrav= 4, a=2.5537, c=6.3985, nat=6, ntyp= 2, ? ecutwfc =35.0, ecutrho=150.0, ? occupations='smearing', smearing='methfessel-paxton', degauss=0.01, ? la2F = .true.,? ?/ ?&electrons ? mixing_beta = 0.7 ? conv_thr = 1.0d-8 ?/ ATOMIC_SPECIES B 10.81000 B.pw91-n-van_ak.UPF? C 12.0107 C.pw91-van_ak.UPF ATOMIC_POSITIONS {crystal} ? B 0.000000000 0.000000000 0.007000000 ? ? C 0.000000000 0.000000000 0.259200000 ? ? C 0.333333300 -0.333333300 0.338600000 ? C 0.333333300 -0.333333300 0.581600000 ? C 0.666666667 -0.666666667 0.664600000 ? C 0.666666667 -0.666666667 0.899900000 ? K_POINTS automatic? 11 11 4 0 0 0 elph. input: ? Electron-phonon coefficients for BC5 ?&inputph ? tr2_ph=1.0d-10, ? prefix='BC5', ? fildvscf='BC5dv', ? amass(1)=10.810, ? amass(2)=12.0107 ? outdir='/Thesis2007/temp' ? fildyn='BC5.dyn', ? elph=.true., ? trans=.true., ? ldisp=.true. ? nq1=2, nq2=2, nq3=3, ?/Last page of elph. output:? Convergence has been achieved? ? ? Number of q in the star = 1 ? List of q in the star: ? 1 0.000000000 0.000000000 0.133036389 ? In addition there is the -q list:? ? 1 0.000000000 0.000000000-0.133036389 ? Diagonalizing the dynamical matrix ? q = ( 0.000000000 0.000000000 0.133036389 )? ?************************************************************************** ? omega( 1) = 5.505487 [THz] = 183.644514 [cm-1] ? omega( 2) = 5.505487 [THz] = 183.644514 [cm-1] ? omega( 3) = 8.997032 [THz] = 300.110679 [cm-1] ? omega( 4) = 10.179023 [THz] = 339.537920 [cm-1] ? omega( 5) = 10.179023 [THz] = 339.537920 [cm-1] ? omega( 6) = 15.641133 [THz] = 521.735520 [cm-1] ? omega( 7) = 15.641133 [THz] = 521.735520 [cm-1] ? omega( 8) = 17.156151 [THz] = 572.271403 [cm-1] ? omega( 9) = 22.199800 [THz] = 740.510529 [cm-1] ? omega(10) = 22.199800 [THz] = 740.510529 [cm-1] ? omega(11) = 26.623991 [THz] = 888.086643 [cm-1] ? omega(12) = 29.688222 [THz] = 990.299072 [cm-1] ? omega(13) = 30.654823 [THz] = 1022.541636 [cm-1] ? omega(14) = 30.654823 [THz] = 1022.541636 [cm-1] ? omega(15) = 32.111991 [THz] = 1071.147834 [cm-1] ? omega(16) = 32.111991 [THz] = 1071.147834 [cm-1] ? omega(17) = 32.709423 [THz] = 1091.076164 [cm-1] ? omega(18) = 34.735445 [THz] = 1158.657439 [cm-1] ?************************************************************************** ? electron-phonon interaction ... ?%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ? from elphsum : error # 3 ? q is not a vector in the dense grid ?%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ? stopping ... Get your preferred Email name! Now you can @ymail.com and @rocketmail.com. http://mail.promotions.yahoo.com/newdomains/aa/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090517/d09285fc/attachment-0001.htm From paulatto at sissa.it Sun May 17 19:16:23 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sun, 17 May 2009 19:16:23 +0200 (CEST) Subject: [Pw_forum] The scf can't converge In-Reply-To: <149263.61872.qm@web15203.mail.cnb.yahoo.com> References: <149263.61872.qm@web15203.mail.cnb.yahoo.com> Message-ID: <56699.78.12.160.143.1242580583.squirrel@webmail.sissa.it> On Sun, May 17, 2009 17:18, wangweiphysics at yahoo.com.cn wrote: > deduce it to 0.01, also failed. You cannot jus reduce it, you have to test convergence with respect to the number of kpoints. I second Axel for the rest, please check your input file with xcrysden. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From eyvaz_isaev at yahoo.com Sun May 17 21:10:05 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 17 May 2009 12:10:05 -0700 (PDT) Subject: [Pw_forum] question about elphonon calculation Message-ID: <65347.7460.qm@web65710.mail.ac4.yahoo.com> Dear Mohazar, > from elphsum : error # 3 > q is not a vector in the dense grid It is (almost) exactly mentioned what is wrong: nq1,nq2,and nq3 in ph.in should be a divisor of the dense grid. > ? ecutwfc =35.0, ecutrho=150.0, ecutrho looks somewhat small, it should be (8-12) times larger than ecutwfs as you use US pseudopotentials. Please provide your affiliation when you post a message to this forum. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com From jasius_1 at yahoo.com Mon May 18 00:03:04 2009 From: jasius_1 at yahoo.com (Jonas Baltrusaitis) Date: Sun, 17 May 2009 15:03:04 -0700 (PDT) Subject: [Pw_forum] monoclinic cell definition question Message-ID: <75444.63577.qm@web54410.mail.yahoo.com> Wow, talk about comprehensive Let me summarize it (mainly so I am sure I understand correctly) say my cif file says" a=2; b=3; c=4; alpha=90; beta=123; gamma=90 So If I move it into pwscf, they would become: a=4; b=2; c=3, alpha=90; beta=90; gamma=123 ADDIONALLY (cif) to (pwscf) x=y y=z z=a the latter is the part I didn't quite get: do we move coords as well? thanks for your help Jonas --- On Wed, 5/13/09, Michael Mehl wrote: > From: Michael Mehl > Subject: Re: [Pw_forum] monoclinic cell definition question > To: jasius_1 at yahoo.com, "PWSCF Forum" > Cc: "Lorenzo Paulatto" > Date: Wednesday, May 13, 2009, 6:32 PM > jasius_1 at yahoo.com > wrote: > > thanks. the problem is that the new gamma between new > a and b is still 90o! > > Sigh.? This is what comes of trying to answer a > question near my > bedtime.? I'm afraid that I've totally fraked the > explanation.? Let's > try again: > > The page http://cst-www.nrl.navy.mil/lattice/struk/g0_6.html > shows a > lattice with unique axis "b", not "c", as is obvious from > the "beta" > angle.? Last night I was thinking it was a "c" > lattice, but like I > said, it was near my bedtime. > > So the problem is: how to convert a monoclinic lattice from > a unique > axis b representation to a unique axis c representation, > and visa > versa, while preserving the shape of the unit cell. > > Apologies to the list, this is going to take a bit of > time: > > By convention, when we say "a monoclinic lattice with > unique axis b" > we mean the second vector of the set of three primitive > vectors is > perpendicular to the other two, and those other two are > not > perpendicular, but are pointed at some non-90 degree angle > relative to > one another. > > So let's look at a set of lattice vectors for this > system.? One choice > (and there are an infinite number) would be: > > a1 = [? ? s? ???, 0 ,? > ? 0? ???] > a2 = [? ? 0? ???, t ,? > ? 0? ???] > a3 = [ u cos(v) , 0 , u sin(v) ] > > where v <> 90 degrees and I deliberately didn't use > a,b,c,alpha,beta,gamma.? Note that this is the choice > made on the web > page in question. > > Now in the traditional expression of lattice vectors: > > a = Sqrt[a1.a1] = s > b = Sqrt[a2.a2] = t > c = Sqrt[a3.a3] = u > > and the angles are defined by > > cos(alpha) = a2.a3/(|a2| |a3|) = 0 > cos(beta)? = a3.a1/(|a3| }a1|) = cos(v) => beta = > v > cos(gamma) = a1.a2/(|a1| |a2|) = 0 > > That's the unique axis b monoclinic lattice: vector b (a2) > is > perpendicular to vectors a (a1) and c (a3). > > Now let's take the same system and express it as a > monoclinic system > with unique axis c.? That is, we want vector a3 to be > perpendicular to > a1 and a2, and a1 and a2 to have the same non-right angle > between them > as vectors a1 and a3 did in the old system, while > preserving the shape > of the primitive cell.? One more or less obvious > choice, which preserves the right-hand nature of the > system, is to just permute (not rotate, duh) the vectors, > and list them as: > > a1' = [ u cos(v) , 0 , u sin(v) ] > a2' = [? ? s? ???, 0 ,? > ? 0? ???] > a3' = [? ? 0? ???, t ,? > ? 0? ???] > > And now: > > a = Sqrt[a1'.a1'] = u > b = Sqrt[a2'.a2'] = s > c = Sqrt[a3'.a3'] = t > > and > > cos(alpha) = a2'.a3'/(|a2'| |a3'|) = 0 > cos(beta)? = a3'.a1'/(|a3'| }a1'|) = 0 > cos(gamma) = a1'.a2'/(|a1'| |a2'|) = cos(v) => gamma = > v > > However (and this is where the "rotate" comment came in), > my personal > quirk is that (a1,a2,a3) should be as close as possible to > (X,Y,Z). > So NOW lets rotate that system about the (111) axis, > taking > Z->X->Y->Z.? We get > > a1" = [ u sin(v) , u cos(v) , 0 ] > a2" = [? ? 0? ???,? > ? s? ???, 0 ] > a3" = [? ? 0? ???,? > ? 0? ???, t ] > > where we leave it as an exercise for the reader to show > that this > is the same combination of (a,b,c,alpha,beta,gamma) as the > primed > vectors above. > > What about the basis of the lattice?? Suppose we have > a basis vector that was written in original crystal (aka > lattice) coordinates as > > B1 = p a1 + q a2 + r a3 > > In Cartesian space, this becomes > > B1 = [p s + r u cos(v), q t , r sin(v) ] > > and a little bit of thought shows that this can be written > in terms of > the primed lattice as: > > B1 = r a1' + p a2' + q a3' > > which is the permutation that someone mentioned. > > If you then want to go to my rotated space, you get > > B1 = r a1" + p a2" + q a3" > ???= [r u sin(v) , r u cos(v) + p s, q t] > > I hope this clears things up, but once again it's getting > near my > bedtime, so it's possible I've made another mistake. > > This is a lot easier to see if you draw some pictures of > the > individual sets of primitive vectors.? You'll see that > all the sets > (a1,a2,a3), (a1',a2',a3') and (a1",a2",a3") define the same > lattice, > and the the corresponding B1 vector is at the same place in > the > primitive cell. > > -- Mike Mehl > Naval Research Laboratory, Washington DC > (At least until someone reads this post) > Disclaimer:? If you've been following this discussion, > you should know > to take everything I've said with a grain of salt. > > From sclauzer at sissa.it Mon May 18 10:12:25 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 18 May 2009 10:12:25 +0200 Subject: [Pw_forum] Hubbard U and small degauss ... In-Reply-To: <97e4b88e0905170301n7f8356b3x9b14c76ed44e4dd2@mail.gmail.com> References: <97e4b88e0905170301n7f8356b3x9b14c76ed44e4dd2@mail.gmail.com> Message-ID: <4A111869.30202@sissa.it> Dear Naol, Naol Regassa wrote: > Dear all > > I have two questions to ask > > a) For the last times I was calculating the effect of Hubbard U on the > system YMnO_3 using QESPRESSO 4.10. I got an insulator band structure > using U = 10.5eV though it is completely metal without U (GGA). When I This is plausible. Though I don't know the physics of your compound, I can tell you that for simpler transition metal oxides (like FeO) you find the same change (i.e. from metal to insulator) after introducing the Hubbard U. Please have a look at Matteo Cococcioni's PhD thesis (from http://www.sissa.it/cm/phd.php) and the related paper (PRB 2005). > got this result I have used small degauss = 0.006. My friend told me > to use degauss = 0.015. Have you asked him the reason? I don't see any... In principle you should test convergence of the total energy (and other quantities you're interested in) w.r.t all the parameters which introduce significant approximations to your result. As you decrease degauss you'll reach the correct result for your system at T=0 (chosing k-point grids accordingly), so I don't see any reason to choose higher degauss other that saving computer time. In the specific case of DFT+U, you have to be carefull to converge the local occupations too. They are very sensitive to the value of ecutwfc, but also degauss and number k-points have to be checked. Please tell me how possible (correct) is > that? I checked the convergence for degauss but it is increasing near > 0.015. What is increasing? > > b) Is adding U value more effective for the most stable structure ? Sorry I cannot get this question... in which sense "more effective"? > For the unstable states like for my case cubic ferromagnetic phase I > can't see any change? What are looking at? Maybe your question is clear to someone who's studying systems like yours, so it's because of me that I cannot get the point... but if you could be more precise maybe you would have more change to get help. Regards, GS > > Here is the input for the 'scf' > &control > calculation = 'scf', > restart_mode = 'from_scratch', > prefix = 'ymo_hcp_ferro' > tstress = .true., > tprnfor = .true., > pseudo_dir = '/home/MSC608/Desktop/QESPRESSO_PP', > outdir='/home/MSC608/Thesis/Hexagonal/Ferro-Hex/temp' > / > &system > ibrav = 0, > celldm(1) = 6.7, > nat = 10, > ntyp = 4, > ecutwfc = 32, > ecutrho = 340.0 > nspin = 2 > starting_magnetization(2) = 0.7, > starting_magnetization(3) = 0.7, > occupations = 'smearing', > smearing = 'marzari-vanderbilt', > degauss = 0.005, > lda_plus_u=.true., > Hubbard_U(2) = 10.5, > Hubbard_U(3) = 10.5, > / > &electrons > diagonalization = 'david' > startingwfc = 'atomic+random' > mixing_beta = 0.3, > conv_thr = 1.0d-10 > / > ATOMIC_SPECIES > Y 88.91 Y.pbe-nsp-van.UPF > Mn1 54.94 Mn.pbe-sp-van_mit.UPF > Mn2 54.94 Mn.pbe-sp-van_mit.UPF > O 15.9994 O.pbe-rrkjus.UPF > > CELL_PARAMETERS (alat) > 0.985236441 -0.266531554 -0.000316189 > -0.266366323 0.985859979 0.000349871 > 0.000781869 -0.000890980 3.186836665 > > ATOMIC_POSITIONS (crystal) > Y -0.082663260 -0.082649328 0.000174846 > Y -0.082874577 -0.082880779 0.499813069 > Mn1 0.205305857 0.560245090 0.249986248 > Mn2 0.561457263 0.207812583 0.750012603 > O -0.027464907 -0.056329933 0.249979343 > O -0.058482388 -0.031289340 0.750035308 > O 0.247951998 0.582250730 0.083800805 > O 0.582338704 0.248188482 0.916046842 > O 0.582135578 0.247948396 0.583964864 > O 0.247742343 0.582041831 0.416186070 > K_POINTS automatic > 3 3 2 1 1 1 > > Thanks alot in advance > Tamene Regassa > AAU > Ethiopia > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From Andrea.Marini at roma2.infn.it Mon May 18 10:09:58 2009 From: Andrea.Marini at roma2.infn.it (Andrea Marini) Date: Mon, 18 May 2009 10:09:58 +0200 (CEST) Subject: [Pw_forum] Electron-Phonon: loading dVscf file when ldisp=.TRUE. Message-ID: Dear all, I am encountering a problem when calculating lambda constants (PH/elephon.f90). Till now I used to 1. calculate first phonons and dVscf file by using elph=.false., trans=.true., ldisp=.false., in the ph.x input file. 2. re-read the dvscf file using a larger k-point grid and changing in the ph.x input file elph=.true., trans=.false., ldisp=.false., This method works great. If I compare the result of point 2 against the calculation 1 using elph=.true. the lambda factors are identical. Now I am trying to follow points 1 and 2 when ldisp=.TRUE.. Here the results of the calculation 2, when the dVscf file is read from disk, are different the case when phonons and lambda's are calculated in the same run. More importantly I get that the symmetry of the lambda factors are not respected for some q-points when the dVscf file is read from disk: < ldisp=.true., trans=.true., elph=.true. > ldisp=.true., trans=.false., elph=.true. < q = ( 0.500000000 -0.500000000 0.500000000 ) < < omega( 1) = 3.751393 [THz] = 125.133833 [cm-1] < omega( 2) = 3.751393 [THz] = 125.133833 [cm-1] < omega( 3) = 9.121952 [THz] = 304.277585 [cm-1] < < lambda( 1)= 2.6170 gamma= 1077.54 GHz < lambda( 2)= 2.6170 gamma= 1077.54 GHz < lambda( 3)= 0.1901 gamma= 462.75 GHz --- > lambda( 1)= 31.9820 gamma=13168.61 GHz > lambda( 2)= 23.5894 gamma= 9712.95 GHz > lambda( 3)= 19.7009 gamma=47963.51 GHz Maybe I am doing something wrong. I would really appreciate is someone could help me. I have attached to this mail a tar.gz file with all inputs to reproduce the above described error in the simple case of Al bulk. All the runs take less than 5 minutes. I have already created all the links so if you want to try, simple > cd Alluminum_TEST/ > cd CALCULATE_DVSCF_LDISP_TRUE/ > pw.x < gs.in > gs.out > ph.x < ph.in > ph.out > cd ../READ_DVSCF_LDISP_TRUE/ > pw.x < gs.in > gs.out > ph.x < ph.in > ph.out > cd CALCULATE_DVSCF_SINGLE_Q/ > pw.x < gs.in > gs.out > pw.x < nscf.in > nscf.out > ph.x < ph.in > ph.out > cd READ_DVSCF_SINGLE_Q > pw.x < gs.in > gs.out > pw.x < nscf.in > nscf.out > ph.x < ph.in > ph.out The diff between the files CALCULATE_DVSCF_LDISP_TRUE/ph.out and READ_DVSCF_LDISP_TRUE/ph.out is saved as DIFF_LDISP_TRUE. Cheers Andrea -------------------------------------------------------------------------- Andrea MARINI Physics Department, University of Rome "Tor Vergata" (Italy) - phone: +39-0672594894 - fax: +39-062023507 - - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - -------------- next part -------------- A non-text attachment was scrubbed... Name: al_test.tar.gz Type: application/x-gzip Size: 23746 bytes Desc: Al bulk TEST Url : http://www.democritos.it/pipermail/pw_forum/attachments/20090518/24c418b2/attachment-0001.bin From panda.deng.pan at gmail.com Mon May 18 10:34:20 2009 From: panda.deng.pan at gmail.com (=?GB2312?B?xcu1xw==?=) Date: Mon, 18 May 2009 16:34:20 +0800 Subject: [Pw_forum] problem about DeltaVscf file missing Message-ID: Dear All Users, I was trying to calculate the el-ph coupling.And I followed just as the example 07 did. First I run a dense scf calculation and then a scf calculation. Sencond I run the phonon calculation and it got error message. I attached the input of mine and the error message below.Although from the error message I thought there should be DV file written down.Can any one help me what I was doing wrong.In addition, the error happened just the ph.x run just started. Please someone give me some advise about this. Have a nice day Deng Pan Nanjing University elph.in MgB2-elph &inputph amass(1)= 24.3, amass(2)= 10.8, prefix='MgB2' tr2_ph=1.0d-14, alpha_mix(1)=0.5, iverbosity=1, fildyn='/MgB2/elph/test/MgB2.dyn', outdir='/tmp1/', ldisp=.true, trans=.true, elph=.true. nq1=4,nq2=4,nq3=4, fildvscf='/MgB2/elph/test/MgB2dv' / and the error message task # 3 from phq_readin : error # 1 El-ph needs a DeltaVscf file -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090518/da409548/attachment.htm From Daniele.Passerone at empa.ch Mon May 18 11:56:05 2009 From: Daniele.Passerone at empa.ch (Passerone, Daniele) Date: Mon, 18 May 2009 11:56:05 +0200 Subject: [Pw_forum] Extracting the kinetic energy from the one-electron contribution Message-ID: Dear users/developers I have a very trivial question, but unfortunately I can not find an immediate solution. How to isolate and print the kinetic energy contribution to the one-electron energy (that is printed at the end of a scf run?). I see in the code that the one-electron contribution is not computed directly as sum of Vion and Ekin, and I suppose that the latter (which is computed in G-space) should be available somewhere... Thank you for any help, Daniele ************************************************************************* Dr. Daniele Passerone Theory and atomistic simulation of materials group Empa Swiss Federal Laboratories for Materials Science and Research nanotech at surfaces laboratory Ueberlandstrasse 129 CH-8600 D?bendorf Switzerland Tel.: +41-(0)44-8234877 Fax : +41-(0)44-8234031 E-mail in case of false blacklisting: daniele.passerone at hispeed.ch From paulatto at sissa.it Mon May 18 12:41:13 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 12:41:13 +0200 (CEST) Subject: [Pw_forum] Extracting the kinetic energy from the one-electron contribution In-Reply-To: References: Message-ID: <35536.78.12.158.147.1242643273.squirrel@webmail.sissa.it> On Mon, May 18, 2009 11:56, Passerone, Daniele wrote: > Dear users/developers > > I have a very trivial question, but unfortunately I can not find an > immediate solution. > How to isolate and print the kinetic energy contribution to the > one-electron energy (that is printed at the end of a scf run?). I'm afraid you question may not be so trivial, the one-electron term (band energy) is the sm of the band eigenvalues multiplied by the band occupancy and integrated over the brillouin zone. The problem is that at each kpoint an hamiltonian (depending on k) is contructed and diagonalized, hence the kinetic term is mixed with the external KS potential. Furthermore, the eigenvalues of the crystal hamiltonian are not eigenstates of the kinetic energy operator, what you can get is the mean value . It does not mean that it is impossible to obtain the kinetic contribution, as it can always be computed from the wavefunctions. Actually, the meta-gga extension to density functionals uses kinetic energy, expressing it in terms of kinetic energy density. With a bit of fiddling you can probably activate the calculation of kinetic energy density without all the other meta-gga stuff; if you grep dft_is_meta through the code you should get some hints. Maybe, if you let us know what you need, instead of what you want, we can find a simpler solution. But if you are really interested in kinetic energy I cannot just tell you "print ekin in kinetic.f90". Best regards, and good luck! -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From Daniele.Passerone at empa.ch Mon May 18 13:40:00 2009 From: Daniele.Passerone at empa.ch (Passerone, Daniele) Date: Mon, 18 May 2009 13:40:00 +0200 Subject: [Pw_forum] Re: Extracting the kinetic energy from the one-electron contribution Message-ID: |Maybe, if you let us know what you need, instead of what you want, we can |find a simpler solution. But if you are really interested in kinetic |energy I cannot just tell you "print ekin in kinetic.f90". I am not sure about the nuance of your request :-) However, I have a defected slab where I observe what I think is bonding/antibonding. At variance with Mott localization this should improve the kinetic energy with respect to a slab where this defect is not present. So probably the kin. energy density, as output of pp.x would be the best of all worlds... then one could integrate over all space, layer by layer, etc... Thanks Daniele ************************************************************************* Dr. Daniele Passerone Theory and atomistic simulation of materials group Empa Swiss Federal Laboratories for Materials Science and Research nanotech at surfaces laboratory Ueberlandstrasse 129 CH-8600 D?bendorf Switzerland Tel.: +41-(0)44-8234877 Fax : +41-(0)44-8234031 E-mail in case of false blacklisting: daniele.passerone at hispeed.ch From paulatto at sissa.it Mon May 18 14:04:44 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 14:04:44 +0200 (CEST) Subject: [Pw_forum] Extracting the kinetic energy from the one-electron contribution In-Reply-To: References: Message-ID: <37455.78.12.158.147.1242648284.squirrel@webmail.sissa.it> On Mon, May 18, 2009 13:40, Passerone, Daniele wrote: > So probably the kin. energy density, as output of pp.x would be the best > of all worlds... Grepping a bit I've found that the postprocessing and totally unrelated utility pw2casino *does* compute the kinetic energy: you can use it to postprocess you data and get the total kinetic energy. If you want it as a density, I think it should be feasible to combine the pw2casino part of the code with the local_dos.f90 code used by pp.x to plot wavefunction-related quantities. The latter has grown a bit messy, though: I can't understand half of what it's doing myself. > then one could integrate over all space, layer by layer, etc... This would be nice, let's wait a bit and see if someone has done it already... I've seen it happening more than once. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From clie at aphy.iphy.ac.cn Mon May 18 14:08:57 2009 From: clie at aphy.iphy.ac.cn (clie) Date: Mon, 18 May 2009 12:08:57 GMT Subject: [Pw_forum] s_psi_nc_ Message-ID: <20090518120857.4998.eqmail@aphy.iphy.ac.cn> Dear friends, I am compiling the QE package,but some error emergences like following libpw.a (s_lpsi.o); in function 's_lpsi_': s_lpsi.f90:(.text+ox1c9) undefined reference to 's_psi_nc_' make: ***[pw.x] Error 2 I don't how to solve, It will be very grateful if some can give some suggestions. All the best, Zhi Li ----------------------------------------------- Ex5 group Extreme Condition Physics, Institute of Physics, Chinese Academy of Sciences, P. O. Box 603, Beijing, 100190, People?s Republic of China From eyvaz_isaev at yahoo.com Mon May 18 14:28:57 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 18 May 2009 05:28:57 -0700 (PDT) Subject: [Pw_forum] lambda.f90 Message-ID: <308730.65667.qm@web65713.mail.ac4.yahoo.com> Dear users, Please update lambda.f90 if you plan to use it to calculate electron-phonon interaction constant \lambda and then evaluate T_c via Allen-Dynes equation. Previous version contains unexpected format error which leads to an error reading dyn-files. I have also slightly changed output format. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com -------------- next part -------------- A non-text attachment was scrubbed... Name: lambda.f90 Type: application/octet-stream Size: 6742 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20090518/804a52c6/attachment-0001.obj From anavat_ila at yahoo.com Mon May 18 14:39:12 2009 From: anavat_ila at yahoo.com (Ali Tavana) Date: Mon, 18 May 2009 05:39:12 -0700 (PDT) Subject: [Pw_forum] k and q points Message-ID: <228960.81420.qm@web51802.mail.re2.yahoo.com> Dear users, I have a simple question: Are the irreducible k or From l.salimi at ph.iut.ac.ir Mon May 18 14:14:21 2009 From: l.salimi at ph.iut.ac.ir (leila salimi) Date: Mon, 18 May 2009 15:44:21 +0330 (IRST) Subject: [Pw_forum] Install PWSCF Message-ID: <28178373.736451242648861836.JavaMail.root@mta.iut.ac.ir> Hi, May I ask you a question? I installed Ifort 9, MKL 9, fftw and MPI correctly, then I tried to install espresso. configure was successful but while "make all" was running, without giving any error message, the installation stoped at this line: mpif90 -O2 -assume byterecl -nomodule -fpp -D__INTEL -D__FFTW -D__MPI -D__PARA -I../include -I/test/leila/FFTW/include -I/test/leila/FFTW/include -I./ -I../Modules -I../iotk/src -I../PW -I../PH -c scf_mod.f90 Do you have any idea what the problem might be? Thanks. Leila From greatnaol at gmail.com Mon May 18 14:48:28 2009 From: greatnaol at gmail.com (Naol Regassa) Date: Mon, 18 May 2009 15:48:28 +0300 Subject: [Pw_forum] Re hubbard U and stable structure Message-ID: <97e4b88e0905180548l3229976eic3f5c0036ee5ce1@mail.gmail.com> Dear GS Thank you for responding to my question. In my question I said "more effective" because I seen some changes on the band structure of cubic ferromagnetic by adding U.But not as effective ( it won't make totally an insulator) as the one I have seen on the (experimentally found) stable structure i.e. hexagonal. On 5/18/09, pw_forum-request at pwscf.org wrote: > Send Pw_forum mailing list submissions to > pw_forum at pwscf.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.democritos.it/mailman/listinfo/pw_forum > or, via email, send a message with subject or body 'help' to > pw_forum-request at pwscf.org > > You can reach the person managing the list at > pw_forum-owner at pwscf.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Pw_forum digest..." > > > Today's Topics: > > 1. Re: Hubbard U and small degauss ... (Gabriele Sclauzero) > 2. Electron-Phonon: loading dVscf file when ldisp=.TRUE. > (Andrea Marini) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 18 May 2009 10:12:25 +0200 > From: Gabriele Sclauzero > Subject: Re: [Pw_forum] Hubbard U and small degauss ... > To: PWSCF Forum > Message-ID: <4A111869.30202 at sissa.it> > Content-Type: text/plain; charset=us-ascii; format=flowed > > Dear Naol, > > Naol Regassa wrote: >> Dear all >> >> I have two questions to ask >> >> a) For the last times I was calculating the effect of Hubbard U on the >> system YMnO_3 using QESPRESSO 4.10. I got an insulator band structure >> using U = 10.5eV though it is completely metal without U (GGA). When I > > This is plausible. Though I don't know the physics of your compound, I can > tell you that > for simpler transition metal oxides (like FeO) you find the same change > (i.e. from metal > to insulator) after introducing the Hubbard U. Please have a look at Matteo > Cococcioni's > PhD thesis (from http://www.sissa.it/cm/phd.php) and the related paper (PRB > 2005). > >> got this result I have used small degauss = 0.006. My friend told me >> to use degauss = 0.015. > > Have you asked him the reason? I don't see any... > > In principle you should test convergence of the total energy (and other > quantities you're > interested in) w.r.t all the parameters which introduce significant > approximations to your > result. As you decrease degauss you'll reach the correct result for your > system at T=0 > (chosing k-point grids accordingly), so I don't see any reason to choose > higher degauss > other that saving computer time. > > In the specific case of DFT+U, you have to be carefull to converge the local > occupations > too. They are very sensitive to the value of ecutwfc, but also degauss and > number k-points > have to be checked. > > Please tell me how possible (correct) is >> that? I checked the convergence for degauss but it is increasing near >> 0.015. > > What is increasing? > > >> >> b) Is adding U value more effective for the most stable structure ? > > Sorry I cannot get this question... in which sense "more effective"? > >> For the unstable states like for my case cubic ferromagnetic phase I >> can't see any change? > > What are looking at? Maybe your question is clear to someone who's studying > systems like > yours, so it's because of me that I cannot get the point... but if you could > be more > precise maybe you would have more change to get help. > > > Regards, > > > GS > > > >> >> Here is the input for the 'scf' >> &control >> calculation = 'scf', >> restart_mode = 'from_scratch', >> prefix = 'ymo_hcp_ferro' >> tstress = .true., >> tprnfor = .true., >> pseudo_dir = '/home/MSC608/Desktop/QESPRESSO_PP', >> outdir='/home/MSC608/Thesis/Hexagonal/Ferro-Hex/temp' >> / >> &system >> ibrav = 0, >> celldm(1) = 6.7, >> nat = 10, >> ntyp = 4, >> ecutwfc = 32, >> ecutrho = 340.0 >> nspin = 2 >> starting_magnetization(2) = 0.7, >> starting_magnetization(3) = 0.7, >> occupations = 'smearing', >> smearing = 'marzari-vanderbilt', >> degauss = 0.005, >> lda_plus_u=.true., >> Hubbard_U(2) = 10.5, >> Hubbard_U(3) = 10.5, >> / >> &electrons >> diagonalization = 'david' >> startingwfc = 'atomic+random' >> mixing_beta = 0.3, >> conv_thr = 1.0d-10 >> / >> ATOMIC_SPECIES >> Y 88.91 Y.pbe-nsp-van.UPF >> Mn1 54.94 Mn.pbe-sp-van_mit.UPF >> Mn2 54.94 Mn.pbe-sp-van_mit.UPF >> O 15.9994 O.pbe-rrkjus.UPF >> >> CELL_PARAMETERS (alat) >> 0.985236441 -0.266531554 -0.000316189 >> -0.266366323 0.985859979 0.000349871 >> 0.000781869 -0.000890980 3.186836665 >> >> ATOMIC_POSITIONS (crystal) >> Y -0.082663260 -0.082649328 0.000174846 >> Y -0.082874577 -0.082880779 0.499813069 >> Mn1 0.205305857 0.560245090 0.249986248 >> Mn2 0.561457263 0.207812583 0.750012603 >> O -0.027464907 -0.056329933 0.249979343 >> O -0.058482388 -0.031289340 0.750035308 >> O 0.247951998 0.582250730 0.083800805 >> O 0.582338704 0.248188482 0.916046842 >> O 0.582135578 0.247948396 0.583964864 >> O 0.247742343 0.582041831 0.416186070 >> K_POINTS automatic >> 3 3 2 1 1 1 >> >> Thanks alot in advance >> Tamene Regassa >> AAU >> Ethiopia >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > > > ------------------------------ > > Message: 2 > Date: Mon, 18 May 2009 10:09:58 +0200 (CEST) > From: Andrea Marini > Subject: [Pw_forum] Electron-Phonon: loading dVscf file when > ldisp=.TRUE. > To: Pw_forum at pwscf.org > Message-ID: > > Content-Type: text/plain; charset="us-ascii" > > Dear all, > > I am encountering a problem when calculating lambda constants > (PH/elephon.f90). Till now I used to > > 1. calculate first phonons and dVscf file by using > > elph=.false., > trans=.true., > ldisp=.false., > > in the ph.x input file. > > 2. re-read the dvscf file using a larger k-point grid and changing in the > ph.x input file > > elph=.true., > trans=.false., > ldisp=.false., > > This method works great. If I compare the result of point 2 against the > calculation 1 using elph=.true. the lambda factors are identical. > > Now I am trying to follow points 1 and 2 when ldisp=.TRUE.. Here the > results of the calculation 2, when the dVscf file is read from disk, are > different the case when phonons and lambda's are calculated in the same > run. > More importantly I get that the symmetry of the lambda factors are not > respected for some q-points when the dVscf file is read from disk: > > < ldisp=.true., trans=.true., elph=.true. >> ldisp=.true., trans=.false., elph=.true. > > < q = ( 0.500000000 -0.500000000 0.500000000 ) > < > < omega( 1) = 3.751393 [THz] = 125.133833 [cm-1] > < omega( 2) = 3.751393 [THz] = 125.133833 [cm-1] > < omega( 3) = 9.121952 [THz] = 304.277585 [cm-1] > < > > < lambda( 1)= 2.6170 gamma= 1077.54 GHz > < lambda( 2)= 2.6170 gamma= 1077.54 GHz > < lambda( 3)= 0.1901 gamma= 462.75 GHz > --- >> lambda( 1)= 31.9820 gamma=13168.61 GHz >> lambda( 2)= 23.5894 gamma= 9712.95 GHz >> lambda( 3)= 19.7009 gamma=47963.51 GHz > > > Maybe I am doing something wrong. I would really appreciate is someone > could help me. > > I have attached to this mail a tar.gz file with all inputs to reproduce > the above described error in the simple case of Al bulk. All the runs take > less than 5 minutes. I have already created all the links so if you want > to try, simple > >> cd Alluminum_TEST/ >> cd CALCULATE_DVSCF_LDISP_TRUE/ >> pw.x < gs.in > gs.out >> ph.x < ph.in > ph.out >> cd ../READ_DVSCF_LDISP_TRUE/ >> pw.x < gs.in > gs.out >> ph.x < ph.in > ph.out >> cd CALCULATE_DVSCF_SINGLE_Q/ >> pw.x < gs.in > gs.out >> pw.x < nscf.in > nscf.out >> ph.x < ph.in > ph.out >> cd READ_DVSCF_SINGLE_Q >> pw.x < gs.in > gs.out >> pw.x < nscf.in > nscf.out >> ph.x < ph.in > ph.out > > The diff between the files CALCULATE_DVSCF_LDISP_TRUE/ph.out and > READ_DVSCF_LDISP_TRUE/ph.out is saved as DIFF_LDISP_TRUE. > > Cheers > > Andrea > > -------------------------------------------------------------------------- > Andrea MARINI > > Physics Department, University of Rome "Tor Vergata" (Italy) > - phone: +39-0672594894 - fax: +39-062023507 - > > - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - > -------------- next part -------------- > A non-text attachment was scrubbed... > Name: al_test.tar.gz > Type: application/x-gzip > Size: 23746 bytes > Desc: Al bulk TEST > Url : > http://www.democritos.it/pipermail/pw_forum/attachments/20090518/24c418b2/attachment.bin > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 23, Issue 43 > **************************************** > From paulatto at sissa.it Mon May 18 14:54:43 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 14:54:43 +0200 (CEST) Subject: [Pw_forum] k and q points In-Reply-To: <228960.81420.qm@web51802.mail.re2.yahoo.com> References: <228960.81420.qm@web51802.mail.re2.yahoo.com> Message-ID: <53002.78.12.158.147.1242651283.squirrel@webmail.sissa.it> On Mon, May 18, 2009 14:39, Ali Tavana wrote: > I have a simple question: Are the irreducible k or The kpoints printed at the beginning of an scf calculation, if automatic kpoints generation is selected, and nosym=.false. (default), ARE irreducible. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From paulatto at sissa.it Mon May 18 14:57:17 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 14:57:17 +0200 (CEST) Subject: [Pw_forum] Install PWSCF In-Reply-To: <28178373.736451242648861836.JavaMail.root@mta.iut.ac.ir> References: <28178373.736451242648861836.JavaMail.root@mta.iut.ac.ir> Message-ID: <53007.78.12.158.147.1242651437.squirrel@webmail.sissa.it> On Mon, May 18, 2009 14:14, leila salimi wrote: > Do you have any idea what the problem might be? Weird, but difficult to solve without further detail. Maybe your PC does not have enough RAM for the compiler to run all its optimization algorithms and it just silently crashes... well provide a bit more of the output and some specific of your system so we can have a look. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From Daniele.Passerone at empa.ch Mon May 18 15:01:10 2009 From: Daniele.Passerone at empa.ch (Passerone, Daniele) Date: Mon, 18 May 2009 15:01:10 +0200 Subject: [Pw_forum] Re: Extracting the kinetic energy from the one-electron contribution Message-ID: |Grepping a bit I've found that the postprocessing and totally unrelated |utility pw2casino *does* compute the kinetic energy: you can use it to |postprocess you data and get the total kinetic energy. Thank you so much Lorenzo! By the way... sometimes my results can be so messy that this utility could be renamed "casino2casino".... :-) Daniele ************************************************************************* Dr. Daniele Passerone Theory and atomistic simulation of materials group Empa Swiss Federal Laboratories for Materials Science and Research nanotech at surfaces laboratory Ueberlandstrasse 129 CH-8600 D?bendorf Switzerland Tel.: +41-(0)44-8234877 Fax : +41-(0)44-8234031 E-mail in case of false blacklisting: daniele.passerone at hispeed.ch From paulatto at sissa.it Mon May 18 15:02:25 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 15:02:25 +0200 (CEST) Subject: [Pw_forum] s_psi_nc_ In-Reply-To: <20090518120857.4998.eqmail@aphy.iphy.ac.cn> References: <20090518120857.4998.eqmail@aphy.iphy.ac.cn> Message-ID: <54761.78.12.158.147.1242651745.squirrel@webmail.sissa.it> On Mon, May 18, 2009 14:08, clie wrote: > Dear friends, > I am compiling the QE package,but some error emergences like following which version of QE are you compiling? If it is a cvs version you may try to rebuild the dependecy tree with "make depend" before compiling. > libpw.a (s_lpsi.o); in function 's_lpsi_': > s_lpsi.f90:(.text+ox1c9) undefined reference to 's_psi_nc_' > make: ***[pw.x] Error 2 The way this function was called has been changed a bit in the last few month but, as far as I know, only in the CVS version regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From sclauzer at sissa.it Mon May 18 15:07:46 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 18 May 2009 15:07:46 +0200 Subject: [Pw_forum] Re hubbard U and stable structure In-Reply-To: <97e4b88e0905180548l3229976eic3f5c0036ee5ce1@mail.gmail.com> References: <97e4b88e0905180548l3229976eic3f5c0036ee5ce1@mail.gmail.com> Message-ID: <4A115DA2.5030303@sissa.it> Naol Regassa wrote: > Dear GS > > Thank you for responding to my question. > In my question I said "more effective" because I seen some changes on > the band structure of cubic ferromagnetic by adding U.But not as > effective ( it won't make totally an insulator) as the one I have seen > on the (experimentally found) stable structure i.e. hexagonal. To figure it out maybe you can have a look at the PDOS in the two cases. The U term changes the position of the Mn states and consequently the bands which come from an hybridization between Mn states and other states. You should understand between which kind of states the gap forms by projecting your bands onto the atomic wavefunctions (use projwfc.x for this). HTH GS > > > On 5/18/09, pw_forum-request at pwscf.org wrote: >> Send Pw_forum mailing list submissions to >> pw_forum at pwscf.org >> >> To subscribe or unsubscribe via the World Wide Web, visit >> http://www.democritos.it/mailman/listinfo/pw_forum >> or, via email, send a message with subject or body 'help' to >> pw_forum-request at pwscf.org >> >> You can reach the person managing the list at >> pw_forum-owner at pwscf.org >> >> When replying, please edit your Subject line so it is more specific >> than "Re: Contents of Pw_forum digest..." >> >> >> Today's Topics: >> >> 1. Re: Hubbard U and small degauss ... (Gabriele Sclauzero) >> 2. Electron-Phonon: loading dVscf file when ldisp=.TRUE. >> (Andrea Marini) >> >> >> ---------------------------------------------------------------------- >> >> Message: 1 >> Date: Mon, 18 May 2009 10:12:25 +0200 >> From: Gabriele Sclauzero >> Subject: Re: [Pw_forum] Hubbard U and small degauss ... >> To: PWSCF Forum >> Message-ID: <4A111869.30202 at sissa.it> >> Content-Type: text/plain; charset=us-ascii; format=flowed >> >> Dear Naol, >> >> Naol Regassa wrote: >>> Dear all >>> >>> I have two questions to ask >>> >>> a) For the last times I was calculating the effect of Hubbard U on the >>> system YMnO_3 using QESPRESSO 4.10. I got an insulator band structure >>> using U = 10.5eV though it is completely metal without U (GGA). When I >> This is plausible. Though I don't know the physics of your compound, I can >> tell you that >> for simpler transition metal oxides (like FeO) you find the same change >> (i.e. from metal >> to insulator) after introducing the Hubbard U. Please have a look at Matteo >> Cococcioni's >> PhD thesis (from http://www.sissa.it/cm/phd.php) and the related paper (PRB >> 2005). >> >>> got this result I have used small degauss = 0.006. My friend told me >>> to use degauss = 0.015. >> Have you asked him the reason? I don't see any... >> >> In principle you should test convergence of the total energy (and other >> quantities you're >> interested in) w.r.t all the parameters which introduce significant >> approximations to your >> result. As you decrease degauss you'll reach the correct result for your >> system at T=0 >> (chosing k-point grids accordingly), so I don't see any reason to choose >> higher degauss >> other that saving computer time. >> >> In the specific case of DFT+U, you have to be carefull to converge the local >> occupations >> too. They are very sensitive to the value of ecutwfc, but also degauss and >> number k-points >> have to be checked. >> >> Please tell me how possible (correct) is >>> that? I checked the convergence for degauss but it is increasing near >>> 0.015. >> What is increasing? >> >> >>> b) Is adding U value more effective for the most stable structure ? >> Sorry I cannot get this question... in which sense "more effective"? >> >>> For the unstable states like for my case cubic ferromagnetic phase I >>> can't see any change? >> What are looking at? Maybe your question is clear to someone who's studying >> systems like >> yours, so it's because of me that I cannot get the point... but if you could >> be more >> precise maybe you would have more change to get help. >> >> >> Regards, >> >> >> GS >> >> >> >>> Here is the input for the 'scf' >>> &control >>> calculation = 'scf', >>> restart_mode = 'from_scratch', >>> prefix = 'ymo_hcp_ferro' >>> tstress = .true., >>> tprnfor = .true., >>> pseudo_dir = '/home/MSC608/Desktop/QESPRESSO_PP', >>> outdir='/home/MSC608/Thesis/Hexagonal/Ferro-Hex/temp' >>> / >>> &system >>> ibrav = 0, >>> celldm(1) = 6.7, >>> nat = 10, >>> ntyp = 4, >>> ecutwfc = 32, >>> ecutrho = 340.0 >>> nspin = 2 >>> starting_magnetization(2) = 0.7, >>> starting_magnetization(3) = 0.7, >>> occupations = 'smearing', >>> smearing = 'marzari-vanderbilt', >>> degauss = 0.005, >>> lda_plus_u=.true., >>> Hubbard_U(2) = 10.5, >>> Hubbard_U(3) = 10.5, >>> / >>> &electrons >>> diagonalization = 'david' >>> startingwfc = 'atomic+random' >>> mixing_beta = 0.3, >>> conv_thr = 1.0d-10 >>> / >>> ATOMIC_SPECIES >>> Y 88.91 Y.pbe-nsp-van.UPF >>> Mn1 54.94 Mn.pbe-sp-van_mit.UPF >>> Mn2 54.94 Mn.pbe-sp-van_mit.UPF >>> O 15.9994 O.pbe-rrkjus.UPF >>> >>> CELL_PARAMETERS (alat) >>> 0.985236441 -0.266531554 -0.000316189 >>> -0.266366323 0.985859979 0.000349871 >>> 0.000781869 -0.000890980 3.186836665 >>> >>> ATOMIC_POSITIONS (crystal) >>> Y -0.082663260 -0.082649328 0.000174846 >>> Y -0.082874577 -0.082880779 0.499813069 >>> Mn1 0.205305857 0.560245090 0.249986248 >>> Mn2 0.561457263 0.207812583 0.750012603 >>> O -0.027464907 -0.056329933 0.249979343 >>> O -0.058482388 -0.031289340 0.750035308 >>> O 0.247951998 0.582250730 0.083800805 >>> O 0.582338704 0.248188482 0.916046842 >>> O 0.582135578 0.247948396 0.583964864 >>> O 0.247742343 0.582041831 0.416186070 >>> K_POINTS automatic >>> 3 3 2 1 1 1 >>> >>> Thanks alot in advance >>> Tamene Regassa >>> AAU >>> Ethiopia >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> -- >> >> >> o ------------------------------------------------ o >> | Gabriele Sclauzero, PhD Student | >> | c/o: SISSA & CNR-INFM Democritos, | >> | via Beirut 2-4, 34014 Trieste (Italy) | >> | email: sclauzer at sissa.it | >> | phone: +39 040 3787 511 | >> | skype: gurlonotturno | >> o ------------------------------------------------ o >> >> >> ------------------------------ >> >> Message: 2 >> Date: Mon, 18 May 2009 10:09:58 +0200 (CEST) >> From: Andrea Marini >> Subject: [Pw_forum] Electron-Phonon: loading dVscf file when >> ldisp=.TRUE. >> To: Pw_forum at pwscf.org >> Message-ID: >> >> Content-Type: text/plain; charset="us-ascii" >> >> Dear all, >> >> I am encountering a problem when calculating lambda constants >> (PH/elephon.f90). Till now I used to >> >> 1. calculate first phonons and dVscf file by using >> >> elph=.false., >> trans=.true., >> ldisp=.false., >> >> in the ph.x input file. >> >> 2. re-read the dvscf file using a larger k-point grid and changing in the >> ph.x input file >> >> elph=.true., >> trans=.false., >> ldisp=.false., >> >> This method works great. If I compare the result of point 2 against the >> calculation 1 using elph=.true. the lambda factors are identical. >> >> Now I am trying to follow points 1 and 2 when ldisp=.TRUE.. Here the >> results of the calculation 2, when the dVscf file is read from disk, are >> different the case when phonons and lambda's are calculated in the same >> run. >> More importantly I get that the symmetry of the lambda factors are not >> respected for some q-points when the dVscf file is read from disk: >> >> < ldisp=.true., trans=.true., elph=.true. >>> ldisp=.true., trans=.false., elph=.true. >> < q = ( 0.500000000 -0.500000000 0.500000000 ) >> < >> < omega( 1) = 3.751393 [THz] = 125.133833 [cm-1] >> < omega( 2) = 3.751393 [THz] = 125.133833 [cm-1] >> < omega( 3) = 9.121952 [THz] = 304.277585 [cm-1] >> < >> >> < lambda( 1)= 2.6170 gamma= 1077.54 GHz >> < lambda( 2)= 2.6170 gamma= 1077.54 GHz >> < lambda( 3)= 0.1901 gamma= 462.75 GHz >> --- >>> lambda( 1)= 31.9820 gamma=13168.61 GHz >>> lambda( 2)= 23.5894 gamma= 9712.95 GHz >>> lambda( 3)= 19.7009 gamma=47963.51 GHz >> >> Maybe I am doing something wrong. I would really appreciate is someone >> could help me. >> >> I have attached to this mail a tar.gz file with all inputs to reproduce >> the above described error in the simple case of Al bulk. All the runs take >> less than 5 minutes. I have already created all the links so if you want >> to try, simple >> >>> cd Alluminum_TEST/ >>> cd CALCULATE_DVSCF_LDISP_TRUE/ >>> pw.x < gs.in > gs.out >>> ph.x < ph.in > ph.out >>> cd ../READ_DVSCF_LDISP_TRUE/ >>> pw.x < gs.in > gs.out >>> ph.x < ph.in > ph.out >>> cd CALCULATE_DVSCF_SINGLE_Q/ >>> pw.x < gs.in > gs.out >>> pw.x < nscf.in > nscf.out >>> ph.x < ph.in > ph.out >>> cd READ_DVSCF_SINGLE_Q >>> pw.x < gs.in > gs.out >>> pw.x < nscf.in > nscf.out >>> ph.x < ph.in > ph.out >> The diff between the files CALCULATE_DVSCF_LDISP_TRUE/ph.out and >> READ_DVSCF_LDISP_TRUE/ph.out is saved as DIFF_LDISP_TRUE. >> >> Cheers >> >> Andrea >> >> -------------------------------------------------------------------------- >> Andrea MARINI >> >> Physics Department, University of Rome "Tor Vergata" (Italy) >> - phone: +39-0672594894 - fax: +39-062023507 - >> >> - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - >> -------------- next part -------------- >> A non-text attachment was scrubbed... >> Name: al_test.tar.gz >> Type: application/x-gzip >> Size: 23746 bytes >> Desc: Al bulk TEST >> Url : >> http://www.democritos.it/pipermail/pw_forum/attachments/20090518/24c418b2/attachment.bin >> >> ------------------------------ >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> >> End of Pw_forum Digest, Vol 23, Issue 43 >> **************************************** >> > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From eyvaz_isaev at yahoo.com Mon May 18 15:08:59 2009 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 18 May 2009 06:08:59 -0700 (PDT) Subject: [Pw_forum] Install PWSCF Message-ID: <598500.30439.qm@web65704.mail.ac4.yahoo.com> Dear Leila, > May I ask you a question? Sure, ... > Do you have any idea what the problem might be? No, no idea until you provide more info about the error message or compilation process. If no complain from the compiler, check your available disk space. May be you have disk full, no space to write any file. Bests, Eyvaz. ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Mon, 5/18/09, leila salimi wrote: > From: leila salimi > Subject: [Pw_forum] Install PWSCF > To: "pw forum" > Date: Monday, May 18, 2009, 4:14 PM > Hi, > May I ask you a question? > I installed Ifort 9, MKL 9, fftw and MPI correctly, > then? I tried to install espresso. configure was > successful but while "make all" was running, without giving > any error message, the installation stoped at this line: > > mpif90 -O2 -assume byterecl -nomodule -fpp -D__INTEL > -D__FFTW -D__MPI -D__PARA -I../include > -I/test/leila/FFTW/include -I/test/leila/FFTW/include > -I./? -I../Modules? -I../iotk/src -I../PW? > -I../PH -c scf_mod.f90 > ? > Do you have any idea what the problem might be? > > Thanks. > Leila > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From clie at aphy.iphy.ac.cn Mon May 18 15:21:37 2009 From: clie at aphy.iphy.ac.cn (clie) Date: Mon, 18 May 2009 13:21:37 GMT Subject: [Pw_forum] s_psi_nc_ (Lorenzo Paulatto)45 In-Reply-To: References: Message-ID: <20090518132138.17358.eqmail@aphy.iphy.ac.cn> Dear Profssor, yes, it is the CVS version. Thank you very much. ----------------------------------------------- Ex5 group Extreme Condition Physics, Institute of Physics, Chinese Academy of Sciences, P. O. Box 603, Beijing, 100190, People?s Republic of China > > On Mon, May 18, 2009 14:08, clie wrote: >> Dear friends, >> I am compiling the QE package,but some error emergences like following > > which version of QE are you compiling? If it is a cvs version you may try > to rebuild the dependecy tree with "make depend" before compiling. > >> libpw.a (s_lpsi.o); in function 's_lpsi_': >> s_lpsi.f90:(.text+ox1c9) undefined reference to 's_psi_nc_' >> make: ***[pw.x] Error 2 > > The way this function was called has been changed a bit in the last few > month but, as far as I know, only in the CVS version > > regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > > > ------------------------------ > > Message: 8 > Date: Mon, 18 May 2009 15:07:46 +0200 > From: Gabriele Sclauzero > Subject: Re: [Pw_forum] Re hubbard U and stable structure > To: PWSCF Forum > Message-ID: <4A115DA2.5030303 at sissa.it> > Content-Type: text/plain; charset=us-ascii; format=flowed > > > > Naol Regassa wrote: >> Dear GS >> >> Thank you for responding to my question. >> In my question I said "more effective" because I seen some changes on >> the band structure of cubic ferromagnetic by adding U.But not as >> effective ( it won't make totally an insulator) as the one I have seen >> on the (experimentally found) stable structure i.e. hexagonal. > > To figure it out maybe you can have a look at the PDOS in the two cases. The U term > changes the position of the Mn states and consequently the bands which come from an > hybridization between Mn states and other states. You should understand between which kind > of states the gap forms by projecting your bands onto the atomic wavefunctions (use > projwfc.x for this). > > HTH > > GS > > >> >> >> On 5/18/09, pw_forum-request at pwscf.org wrote: >>> Send Pw_forum mailing list submissions to >>> pw_forum at pwscf.org >>> >>> To subscribe or unsubscribe via the World Wide Web, visit >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> or, via email, send a message with subject or body 'help' to >>> pw_forum-request at pwscf.org >>> >>> You can reach the person managing the list at >>> pw_forum-owner at pwscf.org >>> >>> When replying, please edit your Subject line so it is more specific >>> than "Re: Contents of Pw_forum digest..." >>> >>> >>> Today's Topics: >>> >>> 1. Re: Hubbard U and small degauss ... (Gabriele Sclauzero) >>> 2. Electron-Phonon: loading dVscf file when ldisp=.TRUE. >>> (Andrea Marini) >>> >>> >>> ---------------------------------------------------------------------- >>> >>> Message: 1 >>> Date: Mon, 18 May 2009 10:12:25 +0200 >>> From: Gabriele Sclauzero >>> Subject: Re: [Pw_forum] Hubbard U and small degauss ... >>> To: PWSCF Forum >>> Message-ID: <4A111869.30202 at sissa.it> >>> Content-Type: text/plain; charset=us-ascii; format=flowed >>> >>> Dear Naol, >>> >>> Naol Regassa wrote: >>>> Dear all >>>> >>>> I have two questions to ask >>>> >>>> a) For the last times I was calculating the effect of Hubbard U on the >>>> system YMnO_3 using QESPRESSO 4.10. I got an insulator band structure >>>> using U = 10.5eV though it is completely metal without U (GGA). When I >>> This is plausible. Though I don't know the physics of your compound, I can >>> tell you that >>> for simpler transition metal oxides (like FeO) you find the same change >>> (i.e. from metal >>> to insulator) after introducing the Hubbard U. Please have a look at Matteo >>> Cococcioni's >>> PhD thesis (from http://www.sissa.it/cm/phd.php) and the related paper (PRB >>> 2005). >>> >>>> got this result I have used small degauss = 0.006. My friend told me >>>> to use degauss = 0.015. >>> Have you asked him the reason? I don't see any... >>> >>> In principle you should test convergence of the total energy (and other >>> quantities you're >>> interested in) w.r.t all the parameters which introduce significant >>> approximations to your >>> result. As you decrease degauss you'll reach the correct result for your >>> system at T=0 >>> (chosing k-point grids accordingly), so I don't see any reason to choose >>> higher degauss >>> other that saving computer time. >>> >>> In the specific case of DFT+U, you have to be carefull to converge the local >>> occupations >>> too. They are very sensitive to the value of ecutwfc, but also degauss and >>> number k-points >>> have to be checked. >>> >>> Please tell me how possible (correct) is >>>> that? I checked the convergence for degauss but it is increasing near >>>> 0.015. >>> What is increasing? >>> >>> >>>> b) Is adding U value more effective for the most stable structure ? >>> Sorry I cannot get this question... in which sense "more effective"? >>> >>>> For the unstable states like for my case cubic ferromagnetic phase I >>>> can't see any change? >>> What are looking at? Maybe your question is clear to someone who's studying >>> systems like >>> yours, so it's because of me that I cannot get the point... but if you could >>> be more >>> precise maybe you would have more change to get help. >>> >>> >>> Regards, >>> >>> >>> GS >>> >>> >>> >>>> Here is the input for the 'scf' >>>> &control >>>> calculation = 'scf', >>>> restart_mode = 'from_scratch', >>>> prefix = 'ymo_hcp_ferro' >>>> tstress = .true., >>>> tprnfor = .true., >>>> pseudo_dir = '/home/MSC608/Desktop/QESPRESSO_PP', >>>> outdir='/home/MSC608/Thesis/Hexagonal/Ferro-Hex/temp' >>>> / >>>> &system >>>> ibrav = 0, >>>> celldm(1) = 6.7, >>>> nat = 10, >>>> ntyp = 4, >>>> ecutwfc = 32, >>>> ecutrho = 340.0 >>>> nspin = 2 >>>> starting_magnetization(2) = 0.7, >>>> starting_magnetization(3) = 0.7, >>>> occupations = 'smearing', >>>> smearing = 'marzari-vanderbilt', >>>> degauss = 0.005, >>>> lda_plus_u=.true., >>>> Hubbard_U(2) = 10.5, >>>> Hubbard_U(3) = 10.5, >>>> / >>>> &electrons >>>> diagonalization = 'david' >>>> startingwfc = 'atomic+random' >>>> mixing_beta = 0.3, >>>> conv_thr = 1.0d-10 >>>> / >>>> ATOMIC_SPECIES >>>> Y 88.91 Y.pbe-nsp-van.UPF >>>> Mn1 54.94 Mn.pbe-sp-van_mit.UPF >>>> Mn2 54.94 Mn.pbe-sp-van_mit.UPF >>>> O 15.9994 O.pbe-rrkjus.UPF >>>> >>>> CELL_PARAMETERS (alat) >>>> 0.985236441 -0.266531554 -0.000316189 >>>> -0.266366323 0.985859979 0.000349871 >>>> 0.000781869 -0.000890980 3.186836665 >>>> >>>> ATOMIC_POSITIONS (crystal) >>>> Y -0.082663260 -0.082649328 0.000174846 >>>> Y -0.082874577 -0.082880779 0.499813069 >>>> Mn1 0.205305857 0.560245090 0.249986248 >>>> Mn2 0.561457263 0.207812583 0.750012603 >>>> O -0.027464907 -0.056329933 0.249979343 >>>> O -0.058482388 -0.031289340 0.750035308 >>>> O 0.247951998 0.582250730 0.083800805 >>>> O 0.582338704 0.248188482 0.916046842 >>>> O 0.582135578 0.247948396 0.583964864 >>>> O 0.247742343 0.582041831 0.416186070 >>>> K_POINTS automatic >>>> 3 3 2 1 1 1 >>>> >>>> Thanks alot in advance >>>> Tamene Regassa >>>> AAU >>>> Ethiopia >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>>> >>> -- >>> >>> >>> o ------------------------------------------------ o >>> | Gabriele Sclauzero, PhD Student | >>> | c/o: SISSA & CNR-INFM Democritos, | >>> | via Beirut 2-4, 34014 Trieste (Italy) | >>> | email: sclauzer at sissa.it | >>> | phone: +39 040 3787 511 | >>> | skype: gurlonotturno | >>> o ------------------------------------------------ o >>> >>> >>> ------------------------------ >>> >>> Message: 2 >>> Date: Mon, 18 May 2009 10:09:58 +0200 (CEST) >>> From: Andrea Marini >>> Subject: [Pw_forum] Electron-Phonon: loading dVscf file when >>> ldisp=.TRUE. >>> To: Pw_forum at pwscf.org >>> Message-ID: >>> >>> Content-Type: text/plain; charset="us-ascii" >>> >>> Dear all, >>> >>> I am encountering a problem when calculating lambda constants >>> (PH/elephon.f90). Till now I used to >>> >>> 1. calculate first phonons and dVscf file by using >>> >>> elph=.false., >>> trans=.true., >>> ldisp=.false., >>> >>> in the ph.x input file. >>> >>> 2. re-read the dvscf file using a larger k-point grid and changing in the >>> ph.x input file >>> >>> elph=.true., >>> trans=.false., >>> ldisp=.false., >>> >>> This method works great. If I compare the result of point 2 against the >>> calculation 1 using elph=.true. the lambda factors are identical. >>> >>> Now I am trying to follow points 1 and 2 when ldisp=.TRUE.. Here the >>> results of the calculation 2, when the dVscf file is read from disk, are >>> different the case when phonons and lambda's are calculated in the same >>> run. >>> More importantly I get that the symmetry of the lambda factors are not >>> respected for some q-points when the dVscf file is read from disk: >>> >>> < ldisp=.true., trans=.true., elph=.true. >>>> ldisp=.true., trans=.false., elph=.true. >>> < q = ( 0.500000000 -0.500000000 0.500000000 ) >>> < >>> < omega( 1) = 3.751393 [THz] = 125.133833 [cm-1] >>> < omega( 2) = 3.751393 [THz] = 125.133833 [cm-1] >>> < omega( 3) = 9.121952 [THz] = 304.277585 [cm-1] >>> < >>> >>> < lambda( 1)= 2.6170 gamma= 1077.54 GHz >>> < lambda( 2)= 2.6170 gamma= 1077.54 GHz >>> < lambda( 3)= 0.1901 gamma= 462.75 GHz >>> --- >>>> lambda( 1)= 31.9820 gamma=13168.61 GHz >>>> lambda( 2)= 23.5894 gamma= 9712.95 GHz >>>> lambda( 3)= 19.7009 gamma=47963.51 GHz >>> >>> Maybe I am doing something wrong. I would really appreciate is someone >>> could help me. >>> >>> I have attached to this mail a tar.gz file with all inputs to reproduce >>> the above described error in the simple case of Al bulk. All the runs take >>> less than 5 minutes. I have already created all the links so if you want >>> to try, simple >>> >>>> cd Alluminum_TEST/ >>>> cd CALCULATE_DVSCF_LDISP_TRUE/ >>>> pw.x < gs.in > gs.out >>>> ph.x < ph.in > ph.out >>>> cd ../READ_DVSCF_LDISP_TRUE/ >>>> pw.x < gs.in > gs.out >>>> ph.x < ph.in > ph.out >>>> cd CALCULATE_DVSCF_SINGLE_Q/ >>>> pw.x < gs.in > gs.out >>>> pw.x < nscf.in > nscf.out >>>> ph.x < ph.in > ph.out >>>> cd READ_DVSCF_SINGLE_Q >>>> pw.x < gs.in > gs.out >>>> pw.x < nscf.in > nscf.out >>>> ph.x < ph.in > ph.out >>> The diff between the files CALCULATE_DVSCF_LDISP_TRUE/ph.out and >>> READ_DVSCF_LDISP_TRUE/ph.out is saved as DIFF_LDISP_TRUE. >>> >>> Cheers >>> >>> Andrea >>> >>> -------------------------------------------------------------------------- >>> Andrea MARINI >>> >>> Physics Department, University of Rome "Tor Vergata" (Italy) >>> - phone: +39-0672594894 - fax: +39-062023507 - >>> >>> - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - >>> -------------- next part -------------- >>> A non-text attachment was scrubbed... >>> Name: al_test.tar.gz >>> Type: application/x-gzip >>> Size: 23746 bytes >>> Desc: Al bulk TEST >>> Url : >>> http://www.democritos.it/pipermail/pw_forum/attachments/20090518/24c418b2/ attachment.bin >>> >>> ------------------------------ >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >>> >>> End of Pw_forum Digest, Vol 23, Issue 43 >>> **************************************** >>> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 23, Issue 45 > **************************************** From paulatto at sissa.it Mon May 18 16:33:16 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 16:33:16 +0200 (CEST) Subject: [Pw_forum] s_psi_nc_ (Lorenzo Paulatto)45 In-Reply-To: <20090518132138.17358.eqmail@aphy.iphy.ac.cn> References: <20090518132138.17358.eqmail@aphy.iphy.ac.cn> Message-ID: <45693.78.12.158.101.1242657196.squirrel@webmail.sissa.it> On Mon, May 18, 2009 15:21, clie wrote: > Dear Profssor, at the moment i'm a graduate student... > yes, it is the CVS version. Thank you very much. I assume rebuilding the dependencies solved the problem, do you confirm? regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From paulatto at sissa.it Mon May 18 20:14:33 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 18 May 2009 20:14:33 +0200 (CEST) Subject: [Pw_forum] Extracting the kinetic energy from the one-electron contribution In-Reply-To: <37455.78.12.158.147.1242648284.squirrel@webmail.sissa.it> References: <37455.78.12.158.147.1242648284.squirrel@webmail.sissa.it> Message-ID: <53061.78.12.158.101.1242670473.squirrel@webmail.sissa.it> On Mon, May 18, 2009 14:04, Lorenzo Paulatto wrote: > On Mon, May 18, 2009 13:40, Passerone, Daniele wrote: > > So probably the kin. energy density, as output of pp.x would be the > > best of all worlds... I report here, for future reference, what we've discussed in private: the function plot_num=8 of pp.x, which plots the Electron Localization Function, actually computes the local density of kinetic energy (PP/elf.f90). It lacks the ultrasoft ionic terms, which could be implemented in the total kinetic energy but are not easily defined in the K density. cheers -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From anvari_meh at physics.iust.ac.ir Tue May 19 14:11:46 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Tue, 19 May 2009 15:41:46 +0330 Subject: [Pw_forum] van der waals force Message-ID: Dear axel Thanks a lot because of your answer. As you said I am trying to understand the DFT-D method. Now I have another question about difference between pseudo potential code & full one. Is the latter suitable for dispersion force. I mean when we use full potential for certain atom is it compatible with the potential of this atom in crystal? By the way I 'll become happy if someone answer my second question in previous mail Best Regards Mehrnaz Anvari Iran University Of Science & Technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090519/1de3fa50/attachment.htm From cristian.degliesposti at unibo.it Tue May 19 14:44:29 2009 From: cristian.degliesposti at unibo.it (Cristian Degli Esposti Boschi) Date: Tue, 19 May 2009 14:44:29 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro Message-ID: <4A12A9AD.3000503@unibo.it> Dear all, I'm having some problems in installing QE 4.0.5 on a Mac Pro, with 2 Intel Xeon Quad Core CPUs, 16 GB RAM, Mac OS X v10.5 Leopard. If I configure with no other options I get the following error: "Unfortunately, this installation of Open MPI was not compiled with Fortran 90 support. As such, the mpif90 compiler is non-functional." The GNU compiler, gfortran in particular, are installed in the system via the FINK package. If I pass the option MPIF90="gfortran", I can make all. However when I try to use the CP code I get "Fortran runtime error: Attempt to allocate negative amount of memory. Possible integer overflow" I reduced considerably the number of the points in the grid for pseudopotentials and then I get "cp.x(32686) malloc: *** mmap(size=546910208) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Operating system error: Cannot allocate memory Out of memory" From the configure log I see that "checking size of int *... 4" Hence, I recalled that for some other applications I had to pass explicitly the flag for 64-bit compilation. So I tried: sudo ./configure MPIF90="gfortran" CFLAGS="-m64" FFLAGS="-m64" F90FLAGS="-m64" CPPFLAGS="-m64" LDFLAGS="-m64" and as a result: "checking size of int *... 8" But, then when I issue sudo make all: "fft_scalar.f90:61.35: INTEGER, PARAMETER :: lwork = 2 * nfftx 1 Error: Symbol 'lwork' at (1) already has basic type of INTEGER" I have seen in the user guide that using 64-bit executables is not particularly wise. Does anyone have suggestions on how to proceed in order to have cp.x working? Do I have to inspect make.sys? Thanks for your time. Cristian -- ___________________________________________________ Cristian Degli Esposti Boschi CNR, CNISM, Unita' di Ricerca di Bologna, c/o Dipartimento di Fisica, Universita' di Bologna viale Berti-Pichat, 6/2, 40127, Bologna, Italia tel. ++39 051 2095114 fax ++39 051 2095113 e-mail: cristian.degliesposti -AT- unibo.it web: http://www.df.unibo.it/fismat/theory ___________________________________________________ From daniel.forrer at unipd.it Tue May 19 15:12:18 2009 From: daniel.forrer at unipd.it (Daniel Forrer) Date: Tue, 19 May 2009 15:12:18 +0200 Subject: [Pw_forum] van der waals force In-Reply-To: References: Message-ID: Dear Mehrnaz, I guess reading this paper could help you. http://www3.interscience.wiley.com/journal/121402787/abstract 2009/5/19 Mehrnaz Anvari > Dear axel > Thanks a lot because of your answer. As you said I am trying to understand > the DFT-D method. Now I have another question about difference between > pseudo potential code & full one. Is the latter suitable for dispersion > force. I mean when we use full potential for certain atom is it compatible > with the potential of this atom in crystal? > By the way I 'll become happy if someone answer my second question in > previous mail > Best Regards > Mehrnaz Anvari > Iran University Of Science & Technology > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ---------------------------------- Daniel Forrer, Ph.D Student Dipartimento di Scienze Chimiche Universit? degli Studi di Padova V. Marzolo 1, Padova mail daniel.forrer at unipd.it ---------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090519/3dfe6ec3/attachment.htm From akohlmey at cmm.chem.upenn.edu Tue May 19 16:51:08 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 19 May 2009 10:51:08 -0400 Subject: [Pw_forum] van der waals force In-Reply-To: References: Message-ID: <7b6913e90905190751v3595f4c7m59c868f7f0120c7b@mail.gmail.com> 2009/5/19 Mehrnaz Anvari : > Dear axel > Thanks a lot because of your answer. As you said I am trying to understand > the DFT-D method. Now I have another question about difference between > pseudo potential code & full one. Is the latter suitable for dispersion dear mehrnaz, this question can be answered with common sense. there are two points to consider: a) pseudopotentials integrate out core electrons. can you imagine those to contribute to dispersion interactions? b) if using pseudopotentials would give you (significantly) different results than an all electron calculation, would it make sense to use pseudopotentials in the first place? > force. I mean when we use full potential for certain atom is it compatible > with the potential of this atom in crystal? a well built and transferable pseudopotential should be behave like the original atom and in principle the code should not care about it, i.e. you can mix and match pseudopotentials with different number of valence electrons for the same element and should get the same structure (unless the semi-core electrons have a significant contribution, e.g. in highly oxidized environments). the problem of insufficient dispersion interactions in DFT manifests itself in the incorrect asymptotic distribution of the density far away from atoms. cheers, axel. > By the way I 'll become happy if someone answer my second question in > previous mail > Best Regards > Mehrnaz Anvari > Iran University Of Science & Technology > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From marzari at MIT.EDU Tue May 19 17:09:28 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Tue, 19 May 2009 11:09:28 -0400 Subject: [Pw_forum] van der waals force In-Reply-To: <7b6913e90905190751v3595f4c7m59c868f7f0120c7b@mail.gmail.com> References: <7b6913e90905190751v3595f4c7m59c868f7f0120c7b@mail.gmail.com> Message-ID: <4A12CBA8.50402@mit.edu> Axel Kohlmeyer wrote: > 2009/5/19 Mehrnaz Anvari : >> Dear axel >> Thanks a lot because of your answer. As you said I am trying to understand >> the DFT-D method. Now I have another question about difference between >> pseudo potential code & full one. Is the latter suitable for dispersion > > dear mehrnaz, > > this question can be answered with common sense. there are > two points to consider: > a) pseudopotentials integrate out core electrons. can you imagine > those to contribute to dispersion interactions? I couldn't, but it was pointed out that Pekka Pykko has done a lot of work discussing the relevance of vdW interactions between filled d-shells (e.g. in silver compounds). nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From akohlmey at cmm.chem.upenn.edu Tue May 19 17:27:06 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 19 May 2009 11:27:06 -0400 Subject: [Pw_forum] van der waals force In-Reply-To: <4A12CBA8.50402@mit.edu> References: <7b6913e90905190751v3595f4c7m59c868f7f0120c7b@mail.gmail.com> <4A12CBA8.50402@mit.edu> Message-ID: <1242746826.10549.407.camel@zero> On Tue, 2009-05-19 at 11:09 -0400, Nicola Marzari wrote: > Axel Kohlmeyer wrote: > > 2009/5/19 Mehrnaz Anvari : > >> Dear axel > >> Thanks a lot because of your answer. As you said I am trying to understand > >> the DFT-D method. Now I have another question about difference between > >> pseudo potential code & full one. Is the latter suitable for dispersion > > > > dear mehrnaz, > > > > this question can be answered with common sense. there are > > two points to consider: > > a) pseudopotentials integrate out core electrons. can you imagine > > those to contribute to dispersion interactions? > > > I couldn't, but it was pointed out that Pekka Pykko has done a lot > of work discussing the relevance of vdW interactions between filled > d-shells (e.g. in silver compounds). interesting point. i would argue that in this case a pseudopotential that does not include those d-states in the valence would be inadequate to begin with, but - of course - this is a bit of a circular argument... ;) cheers, axel. > > nicola > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From sclauzer at sissa.it Tue May 19 17:53:28 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 19 May 2009 17:53:28 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro In-Reply-To: <4A12A9AD.3000503@unibo.it> References: <4A12A9AD.3000503@unibo.it> Message-ID: <4A12D5F8.4020905@sissa.it> Dear Cristian, Cristian Degli Esposti Boschi wrote: > Dear all, > > I'm having some problems in installing QE 4.0.5 on a > Mac Pro, with 2 Intel Xeon Quad Core CPUs, 16 GB RAM, > Mac OS X v10.5 Leopard. If I configure with no other options > I get the following error: > > "Unfortunately, this installation of Open MPI was not compiled with > Fortran 90 support. As such, the mpif90 compiler is non-functional." OK, that's common for the 10.5 Leopard... they put a mpif90 command in /usr/bin even if it's not usable, since f90 compiler is not installed by default. > > The GNU compiler, gfortran in particular, are installed in the > system via the FINK package. Have you also compiled a MPI version on top of gfortran? > If I pass the option MPIF90="gfortran", I can make all. That sounds strange... even if you build a mpif90 on top of gfortran (for instance using OpenMPI), you should call it as "mpif90", not "gfortran". I don't understand if you want to compile a serial or a parallel version. If you want a serial one, use ./configure --disable-parallel Otherwise, if you want a parallel one I suggest you to compile openmpi on top of ifort or gfortran and then export the correct path for mpif90 to the shell (e.g. export PATH=/path/to/your/mpif90:$PATH) or simply set MPIF90 with the full path to the mpif90 you want to use. Remember also to set the correct library path, otherwise the mpi linker may not find the ones corresponding to the compiler. BTW, I think it is not good to substitute the /usr/bin/mpif90 (which actually is not usable in the default Leopard installation) with your compiled version, but rather to tell the make to use your version. > However when I try to use the CP code I get > > "Fortran runtime error: Attempt to allocate negative amount of memory. > Possible integer overflow" Do the other programs of the QE suite work (for instance pw.x)? I compiled serially on Leopard with Intel CPU using gfortran, g95 and ifort and pw.x seems to work fine (just don't use the ATLAS libs coming with the Leopard bundle, which can give problems, see here: http://www.quantum-espresso.org/wiki/index.php/Installation#Intel_Mac_OS_X_10.4_and_10.5_with_g95_and_gfortran ). I also compiled a parallel version with openmpi on top of ifort and seems to work fine too. I never tried with cp.x so far, but I may if you tell me that pw.x works also for you. From your error messages here below it seems that you're running in some other kind of issues, so maybe my suggestions are not very useful... GS > > I reduced considerably the number of the points in the grid > for pseudopotentials and then I get > > "cp.x(32686) malloc: *** mmap(size=546910208) failed (error code=12) > *** error: can't allocate region > *** set a breakpoint in malloc_error_break to debug > Operating system error: Cannot allocate memory > Out of memory" > > From the configure log I see that > > "checking size of int *... 4" > > > Hence, I recalled that for some other applications I had to pass > explicitly the flag for 64-bit compilation. So I tried: > > sudo ./configure MPIF90="gfortran" CFLAGS="-m64" FFLAGS="-m64" > F90FLAGS="-m64" CPPFLAGS="-m64" LDFLAGS="-m64" > > and as a result: > > "checking size of int *... 8" > > But, then when I issue sudo make all: > > "fft_scalar.f90:61.35: > > INTEGER, PARAMETER :: lwork = 2 * nfftx > 1 > Error: Symbol 'lwork' at (1) already has basic type of INTEGER" > > > I have seen in the user guide that using 64-bit executables > is not particularly wise. Does anyone have suggestions on > how to proceed in order to have cp.x working? Do I have to > inspect make.sys? > > Thanks for your time. Cristian -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sunyang198498 at gmail.com Wed May 20 17:06:32 2009 From: sunyang198498 at gmail.com (=?GB2312?B?y+/R9A==?=) Date: Wed, 20 May 2009 10:06:32 -0500 Subject: [Pw_forum] Questions about nscf calculation Message-ID: <98e7a0810905200806i19151ebcm7770e8b64f10632d@mail.gmail.com> Dear all users: I have a problem regarding the non-selfconsistent calculation. My system involves a metallic slab under an external electric filed with a magnitude of 1.0V/A, equivalently about 0.02 au. the self consistent calculation was OK, and I compare the electric field with the screening charges. It gives me a satiesfied result. Then in order to extract the local density of states, I did the non-self consistent calculation with fine K-point mesh, and then do the local density of states.However, It gives quite astonishing result. Not only the shape of the local density of states was significantly changed from one to the other on the edges of the slab, but also overestimated the screening charges. I went back and used the wave functions generated by the scf calculations to do the local density of states. It also have a huge difference with the one using the nscf calculation. I wonder where the problem is. Following is the input for nscf calculation. &control calculation='nscf' restart_mode='from_scratch', wf_collect = .true. pseudo_dir = '$PSEUDO_DIR/', outdir='$TMP_DIR/' prefix='Pd' / &system ibrav = 6, celldm(1) = 5.18564536158 celldm(3) = 8.48528137424 nat = 7 nspin = 2 starting_magnetization = 0.5 ntyp = 1 ecutwfc = 100.0, ecutrho = 800.0, occupations='smearing', degauss= 0.01 edir = 3 emaxpos=0.4 eopreg=0.2 eamp = -0.01944 nosym = .true. / &electrons conv_thr = 1.0e-8 mixing_beta = 0.7 startingpot='file' startingwfc='atomic' mixing_mode='local-TF' mixing_beta = 0.1 mixing_ndim = 12 / &ions / ATOMIC_SPECIES Pd 1.0 Pd.pz-nd-rrkjus.UPF ATOMIC_POSITIONS (alat) Pd 0.500000000 0.500000000 -2.107651139 Pd 0.000000000 0.000000000 -1.412052059 Pd 0.500000000 0.500000000 -0.706700124 Pd 0.000000000 0.000000000 0.000000010 Pd 0.500000000 0.500000000 0.706700124 Pd 0.000000000 0.000000000 1.412052059 Pd 0.500000000 0.500000000 2.107651139 K_POINTS (automatic) 16 16 1 0 0 0 EOF echo echo "#######################################################################" echo Running PW.x for the nscf calculation ... $PW_COMMAND < nscf.in > nscf.out echo done -- University of Nebraska at Lincoln From dalcorso at sissa.it Wed May 20 18:25:29 2009 From: dalcorso at sissa.it (Dal Corso Andrea) Date: Wed, 20 May 2009 18:25:29 +0200 Subject: [Pw_forum] Electron-Phonon: loading dVscf file when ldisp=.TRUE. In-Reply-To: References: Message-ID: <1242836729.3277.14.camel@dhpc-5-03.sissa.it> On Mon, 2009-05-18 at 10:09 +0200, Andrea Marini wrote: > Dear all, > > I am encountering a problem when calculating lambda constants > (PH/elephon.f90). Till now I used to > > 1. calculate first phonons and dVscf file by using > > elph=.false., > trans=.true., > ldisp=.false., > > in the ph.x input file. > > 2. re-read the dvscf file using a larger k-point grid and changing in the > ph.x input file > > elph=.true., > trans=.false., > ldisp=.false., > > This method works great. If I compare the result of point 2 against the > calculation 1 using elph=.true. the lambda factors are identical. > > Now I am trying to follow points 1 and 2 when ldisp=.TRUE.. Here the > results of the calculation 2, when the dVscf file is read from disk, are > different the case when phonons and lambda's are calculated in the same > run. Presently the ph.x code saves on disk only the information on the actual q point. So after a ldisp=.true. calculation you have on disk only the dVscf of the last q point and you cannot calculate the elph for the other q points. I have right now committed a possible fix in the CVS version. HTH Andrea > More importantly I get that the symmetry of the lambda factors are not > respected for some q-points when the dVscf file is read from disk: > > < ldisp=.true., trans=.true., elph=.true. > > ldisp=.true., trans=.false., elph=.true. > > < q = ( 0.500000000 -0.500000000 0.500000000 ) > < > < omega( 1) = 3.751393 [THz] = 125.133833 [cm-1] > < omega( 2) = 3.751393 [THz] = 125.133833 [cm-1] > < omega( 3) = 9.121952 [THz] = 304.277585 [cm-1] > < > > < lambda( 1)= 2.6170 gamma= 1077.54 GHz > < lambda( 2)= 2.6170 gamma= 1077.54 GHz > < lambda( 3)= 0.1901 gamma= 462.75 GHz > --- > > lambda( 1)= 31.9820 gamma=13168.61 GHz > > lambda( 2)= 23.5894 gamma= 9712.95 GHz > > lambda( 3)= 19.7009 gamma=47963.51 GHz > > > Maybe I am doing something wrong. I would really appreciate is someone > could help me. > > I have attached to this mail a tar.gz file with all inputs to reproduce > the above described error in the simple case of Al bulk. All the runs take > less than 5 minutes. I have already created all the links so if you want > to try, simple > > > cd Alluminum_TEST/ > > cd CALCULATE_DVSCF_LDISP_TRUE/ > > pw.x < gs.in > gs.out > > ph.x < ph.in > ph.out > > cd ../READ_DVSCF_LDISP_TRUE/ > > pw.x < gs.in > gs.out > > ph.x < ph.in > ph.out > > cd CALCULATE_DVSCF_SINGLE_Q/ > > pw.x < gs.in > gs.out > > pw.x < nscf.in > nscf.out > > ph.x < ph.in > ph.out > > cd READ_DVSCF_SINGLE_Q > > pw.x < gs.in > gs.out > > pw.x < nscf.in > nscf.out > > ph.x < ph.in > ph.out > > The diff between the files CALCULATE_DVSCF_LDISP_TRUE/ph.out and > READ_DVSCF_LDISP_TRUE/ph.out is saved as DIFF_LDISP_TRUE. > > Cheers > > Andrea > > -------------------------------------------------------------------------- > Andrea MARINI > > Physics Department, University of Rome "Tor Vergata" (Italy) > - phone: +39-0672594894 - fax: +39-062023507 - > > - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - > _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From Andrea.Marini at roma2.infn.it Wed May 20 18:58:59 2009 From: Andrea.Marini at roma2.infn.it (Andrea Marini) Date: Wed, 20 May 2009 18:58:59 +0200 (CEST) Subject: [Pw_forum] Electron-Phonon: loading dVscf file when ldisp=.TRUE. In-Reply-To: <1242836729.3277.14.camel@dhpc-5-03.sissa.it> References: <1242836729.3277.14.camel@dhpc-5-03.sissa.it> Message-ID: On Wed, 20 May 2009, Dal Corso Andrea wrote: >> Now I am trying to follow points 1 and 2 when ldisp=.TRUE.. Here the >> results of the calculation 2, when the dVscf file is read from disk, are >> different the case when phonons and lambda's are calculated in the same >> run. > > Presently the ph.x code saves on disk only the information on the actual > q point. So after a ldisp=.true. calculation you have on disk only the > dVscf of the last q point and you cannot calculate the elph for the > other q points. > I have right now committed a possible fix in the CVS version. Dear Andrea, many many thanks ! I will try the CVS source ASAP and I will post the results of my tests. Cheers Andrea -------------------------------------------------------------------------- Andrea MARINI Physics Department, University of Rome "Tor Vergata" (Italy) - phone: +39-0672594894 - fax: +39-062023507 - - andrea.marini at roma2.infn.it -- http://www.yambo-code.org/people/andrea - From ddrsmile at gmail.com Thu May 21 05:08:08 2009 From: ddrsmile at gmail.com (DdrSmiLe) Date: Thu, 21 May 2009 11:08:08 +0800 Subject: [Pw_forum] Question about wavefunction, reciporcal lattice and elphon matrix Message-ID: Hi all I want to print out the wavefunction and its corresponding reciprocal lattice vector. And elphon matrix index which contents elphon matrix and its i,j,k,q info. Does anyone have any suggestion or idea? Which f90 code can I modify? How do I modify it? Thank you very much IAMS, Academic Sinica Research Assistant Joey, Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090521/05339d73/attachment.htm From paulatto at sissa.it Thu May 21 09:23:18 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 21 May 2009 09:23:18 +0200 (CEST) Subject: [Pw_forum] Question about wavefunction, reciporcal lattice and elphon matrix In-Reply-To: References: Message-ID: <51136.78.12.163.42.1242890598.squirrel@webmail.sissa.it> On Thu, May 21, 2009 05:08, DdrSmiLe wrote: > Does anyone have any suggestion or idea? You can already do it with pp.x, please read the guide Doc/INPUT_PP.html for details. Unless you take the wavefunctions from the Gamma point, wavefunctions are complex, hence the square module is saved. Nevertheless, it is trivial to modify the code (PP/post_proc.f90) to make it save the real and imaginary part. At the moment (and probably forever) you can only extract wavefunctions for kpoint (what you call reciprocal vectors) that where included in the self-consistent calculation. If you want an arbitrary kpoint you have to pregenerate the wavefuntions doing a non-self-consistent calculation. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From ddrsmile at gmail.com Thu May 21 11:34:53 2009 From: ddrsmile at gmail.com (DdrSmiLe) Date: Thu, 21 May 2009 17:34:53 +0800 Subject: [Pw_forum] Question about wavefunction, reciporcal lattice and elphon matrix In-Reply-To: <51136.78.12.163.42.1242890598.squirrel@webmail.sissa.it> References: <51136.78.12.163.42.1242890598.squirrel@webmail.sissa.it> Message-ID: Dear Lorenzo Paulatto Thank you for your quick reply. I have read INPUT_PP.html. I don't know which plot_num can show wavefunction or how to set the input file to make it generate the files content info of wavefunction and kpoints My purpose is to print out the info of wavefunction, k points and elph_matrix indexes after pw.x or ph.x are completed. Therefore, I think I have to know which matrixes collectes these info. Then I can simply modify the codes to achieve my purpose. So far, I print out the elph_matrix by modifying elphon.f90 elphmat(j,i)= ipert= 1 ibnd= 1 ig evc(complex, real img) 1 0.6683811960749346 -0.4860617886287616 2 0.2651733872655152 -0.1928400315088636 3 0.2651733872655152 -0.1928400315088628 4 -0.0005605875763412 0.0004076718519918 5 -0.0005605875763366 0.0004076718519851 6 -0.0005605875763413 0.0004076718519924 . . . My question is what the ipert, psi and ig mean. Can I list the elphon matrix following (k+q,j) and (k, i) ? Thank you for your help IAMS, Academic Sinica Research Assistant 2009/5/21 Lorenzo Paulatto > > On Thu, May 21, 2009 05:08, DdrSmiLe wrote: > > Does anyone have any suggestion or idea? > > You can already do it with pp.x, please read the guide Doc/INPUT_PP.html > for details. Unless you take the wavefunctions from the Gamma point, > wavefunctions are complex, hence the square module is saved. Nevertheless, > it is trivial to modify the code (PP/post_proc.f90) to make it save the > real and imaginary part. > > At the moment (and probably forever) you can only extract wavefunctions > for kpoint (what you call reciprocal vectors) that where included in the > self-consistent calculation. If you want an arbitrary kpoint you have to > pregenerate the wavefuntions doing a non-self-consistent calculation. > > best regards > > -- > Lorenzo Paulatto > SISSA & DEMOCRITOS (Trieste) > phone: +39 040 3787 511 > skype: paulatz > www: http://people.sissa.it/~paulatto/ > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by SquirrelMail http://www.squirrelmail.org/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090521/b4180935/attachment.htm From paulatto at sissa.it Thu May 21 15:56:42 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 21 May 2009 15:56:42 +0200 Subject: [Pw_forum] Question about wavefunction, reciporcal lattice and elphon matrix In-Reply-To: References: <51136.78.12.163.42.1242890598.squirrel@webmail.sissa.it> Message-ID: In data 21 maggio 2009 alle ore 11:34:53, DdrSmiLe ha scritto: > I have read INPUT_PP.html. > I don't know which plot_num can show wavefunction It's plot number 7, |psi|^2 means squre modulus of wavefunction... > My purpose is to print out the info of wavefunction, k points and > elph_matrix indexes after pw.x or ph.x are completed. Why do you want to do that? Wavefunctions are not very significative quantities, they are tridimensional complex fields: not very easy to visualize. > Therefore, I think I have to know which matrixes collectes these info. > Then I can simply modify the codes to achieve my purpose. Do you want to know the names of the arrays that store the fourier transforms of the wavefunctions? It is evc, from the wavefunctions_module module (Modules/wavefunctions.f90). > My question is what the ipert, psi and ig mean. From what I can see: ipert is an index of perturbation in the specific irreducible representation (whatever it means), psi is the wavefunction (real and complex part), ig is the index of a point in the reciprocal space. > Can I list the elphon matrix following (k+q,j) and (k, i) ? I'm not sure I understand correctly what you are asking. I fail to see what could prevent you from doing it! regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From dimpy.sharma at tyndall.ie Thu May 21 20:21:01 2009 From: dimpy.sharma at tyndall.ie (Dimpy Sharma) Date: Thu, 21 May 2009 19:21:01 +0100 Subject: [Pw_forum] K points in scf calculation Message-ID: Hi all, I am doing some scf calculations in quantum espresso, actually what I want to know is about k-points . I mean when we are doing calculations for a crystal, can we use k point crystal instead of k point automatic? As from the manual of Quantum espresso, what I found is that k point crystal reads the k points in crystal coordinates, i.e in relative coordiantes of the reciprocal lattice lattice vectors while k point automatic generates list of k points. I could not follow this difference properly, can anyone please help me . Thanks Dimpy -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090521/b4eb9a77/attachment.htm From akohlmey at cmm.chem.upenn.edu Thu May 21 20:49:04 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 21 May 2009 14:49:04 -0400 Subject: [Pw_forum] K points in scf calculation In-Reply-To: References: Message-ID: <1242931744.10549.2802.camel@zero> On Thu, 2009-05-21 at 19:21 +0100, Dimpy Sharma wrote: > > Hi all, > > I am doing some scf calculations in quantum espresso, actually what I > want to know is about k-points . I mean when we are doing calculations > for a crystal, can we use k point crystal instead of k point > automatic? As from the manual of Quantum espresso, what I found is yes. why not? > that k point crystal reads the > k points in crystal coordinates, i.e in relative coordiantes of the > reciprocal lattice lattice vectors while k point automatic generates > list of k points. I could not follow this difference properly, can > anyone please help me . the main difference is that with "automatic" you specify what kind of grid you want and QE will create the grid points for you, whereas with all other ways you have to specify the k-points explicitly. cheers, axel. > > Thanks > > Dimpy > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From vranjan at ncsu.edu Thu May 21 22:51:15 2009 From: vranjan at ncsu.edu (Vivek Ranjan) Date: Thu, 21 May 2009 16:51:15 -0400 (EDT) Subject: [Pw_forum] CONSTRAINT card Message-ID: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> Hello, I am wondering how to use CONSTRAINT card ? I want to keep a dihedral angle fixed while letting others relax. The way I do it is the following: I add the following lines: CONSTRAINTS 1 'torsional_angle' 6 7 8 13 150 at the end of the input file, hoping the program will keep the dihedral angle fixed to 150 degrees. However, there is no hint in the output if the above input lines are read, and if the program tries to maintain that angle. After several steps of atomic relaxation, final relaxed structure is calculated. When I check the dihedral angle of the final structure using xcrysden, I find that all the dihedral angles are changed and none of them are equal to 150 degrees. Is there any help ? -- Thank you and Regards, Vivek Ranjan Physics Department NC State University, Raleigh, NC From kboryse at ncsu.edu Fri May 22 07:26:54 2009 From: kboryse at ncsu.edu (Kostyantyn Borysenko) Date: Fri, 22 May 2009 01:26:54 -0400 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back Message-ID: Hi everyone, I'm having problems with the math behind some subroutines. I am talking about PH/trntnsc.f90 and PW/cryst_to_car.f90. These subroutines are supposed to convert vectors (cryst_to_car.f90) or matrices (trntnsc.f90) from crystal coordinates to Cartesian (flag=1) or vice versa (flag=-1). And it seems like a very straightforward task - just a little bit of vector algebra. But when I derive these transformations myself they don't check out. Here is what I've got. By definition of reciprocal vectors we have the following property for dot-product: (a(i),b(j))= 2*pi*delta(i,j), (1a) where a(i) - lattice vectors, b(j) - reciprocal lattice vectors. In QE these vectors are stored in matrices at(i,j) and bg(i,j) respectively. Matrix elements of at are in units a_0 and matrix elements of bg are in units (2*pi/a_0). So the condition (1a) can be rewritten as follows: SUM(k){at(k,i)*bg(k,j)} = delta(i,j). (1b) In matrix form it can be represented as B' * A = I, or B' = inv(A), (2) where B' is a transposed matrix B and inv(A) is an inversed matrix A, and I is the identity matrix. I checked these matrices for the case of graphene and property (2) indeed holds true. Next, it's easy to show that transformation of an arbitrary vector v from crystal to Cartesian coordinates is described by the matrix equation: v_cart = A*v_cryst, (3a) where A is a matrix at(i,j) mentioned above, which contains lattice vector components. If you multiply equation (3a) from the left by inv(A) and use the property inv(A)=B' (see eq. (2)) you will get the reverse transformation: v_cryst = B' * v_cart (4a) Obviously, for an arbitrary matrix W the transformation equations will be W_cart = A' *W_cryst * A (3b) W_cryst = B * W_cart * B' (4b) But that's not what I see in the code. In case of subroutine trntnsc.f90 matrices A and B switch places in (3b) and (4b). And in cryst_to_car.f90 everything seems fine but only if you use at(i,j) as a transformation matrix when flag=1 and bg(i,j) if flag=-1. However, when it's called in subroutine lint.f90 it looks like this: call cryst_to_cart (nkh,xp,at,-1) Like I said, I think the correct result can be obtained only in two cases: call lint(nkh,xp,at,+1) - Crystal -> Cartesian call lint(nkh,xp,bg,-1) - Cartesian -> Crystal Maybe I missed something but this seems like a pretty simple problem from vector algebra. Any ideas? Best regards, Kostyantyn Borysenko Department of Electrical and Computer Engineering North Carolina State University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090522/d4d74362/attachment-0001.htm From sclauzer at sissa.it Fri May 22 08:55:44 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 22 May 2009 08:55:44 +0200 Subject: [Pw_forum] CONSTRAINT card In-Reply-To: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> References: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> Message-ID: <4A164C70.6050408@sissa.it> Please include the full input file in your post (or at least the relevant fraction of it), otherwise it is difficult to understand if you are using correctly this feature of pw. GS Vivek Ranjan wrote: > Hello, > > I am wondering how to use CONSTRAINT card ? I want to keep a dihedral > angle fixed while letting others relax. The way I do it is the > following: > > I add the following lines: > > CONSTRAINTS > 1 > 'torsional_angle' 6 7 8 13 150 > > > at the end of the input file, hoping the program will keep the > dihedral angle fixed to 150 degrees. > > However, there is no hint in the output if the above input lines are > read, and if the program tries to maintain that angle. After several > steps of atomic relaxation, final relaxed structure is calculated. > When I check the dihedral angle of the final structure using xcrysden, > I find that all the dihedral angles are changed and none of them are > equal to 150 degrees. > > Is there any help ? > > > -- > Thank you and Regards, > > Vivek Ranjan > Physics Department > NC State University, Raleigh, NC > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From paulatto at sissa.it Fri May 22 10:11:00 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Fri, 22 May 2009 10:11:00 +0200 (CEST) Subject: [Pw_forum] CONSTRAINT card In-Reply-To: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> References: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> Message-ID: <49357.78.12.167.137.1242979860.squirrel@webmail.sissa.it> On Thu, May 21, 2009 22:51, Vivek Ranjan wrote: > However, there is no hint in the output if the above input lines are > read, They are, for some reason parsing of constrain does not produce any output feedback. I've added it in my local copy of QE but never put it in the CVS. I can send it to you later, if you wish. > I find that all the dihedral angles are changed and none of them are > equal to 150 degrees. There is a tolerance parameter that is set globally for all the constraints that you can specify, it is not very clear to me which unit of measure is used to specify the tolerance in the case of constraints on angles, but you can try and reduce it. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From giuseppe.mattioli at mlib.ism.cnr.it Fri May 22 10:46:14 2009 From: giuseppe.mattioli at mlib.ism.cnr.it (Giuseppe Mattioli) Date: Fri, 22 May 2009 10:46:14 +0200 Subject: [Pw_forum] K points in scf calculation In-Reply-To: <1242931744.10549.2802.camel@zero> References: <1242931744.10549.2802.camel@zero> Message-ID: <200905221046.15440.giuseppe.mattioli@mlib.ism.cnr.it> In other words, either you know very well what you need, and then you write by hand all of the k-points in crystal coordinates, or you use an automatic tool to generate a balanced grid of k-points. Remember only that: 1 - in the former case, if you want to use ONLY the k-points you write down, then you must (if I remember well) switch on the keyword nosym=.true. 2 - in the latter case, the total number of k-points used depends upon the symmetry of your system, so make some tests in order to check whether you can afford the computational cost... On Thursday 21 May 2009 20:49:04 Axel Kohlmeyer wrote: > On Thu, 2009-05-21 at 19:21 +0100, Dimpy Sharma wrote: > > Hi all, > > > > I am doing some scf calculations in quantum espresso, actually what I > > want to know is about k-points . I mean when we are doing calculations > > for a crystal, can we use k point crystal instead of k point > > automatic? As from the manual of Quantum espresso, what I found is > > yes. why not? > > > that k point crystal reads the > > k points in crystal coordinates, i.e in relative coordiantes of the > > reciprocal lattice lattice vectors while k point automatic generates > > list of k points. I could not follow this difference properly, can > > anyone please help me . > > the main difference is that with "automatic" you specify > what kind of grid you want and QE will create the grid points > for you, whereas with all other ways you have to specify the > k-points explicitly. > > cheers, > axel. > > > Thanks > > > > Dimpy > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum -- ******************************************************** - Article premier - Les hommes naissent et demeurent libres et ?gaux en droits. Les distinctions sociales ne peuvent ?tre fond?es que sur l'utilit? commune - Article 2 - Le but de toute association politique est la conservation des droits naturels et imprescriptibles de l'homme. Ces droits sont la libert?, la propri?t?, la s?ret? et la r?sistance ? l'oppression. ******************************************************** Giuseppe Mattioli CNR - ISTITUTO DI STRUTTURA DELLA MATERIA v. Salaria Km 29,300 - C.P. 10 I 00016 - Monterotondo Stazione (RM) Tel + 39 06 90672836 - Fax +39 06 90672316 E-mail: -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090522/72a7a7dd/attachment.htm From dalcorso at sissa.it Fri May 22 11:34:32 2009 From: dalcorso at sissa.it (Dal Corso Andrea) Date: Fri, 22 May 2009 11:34:32 +0200 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back In-Reply-To: References: Message-ID: <1242984872.3251.5.camel@dhpc-5-03.sissa.it> On Fri, 2009-05-22 at 01:26 -0400, Kostyantyn Borysenko wrote: > Hi everyone, > > I'm having problems with the math behind some subroutines. I am > talking about PH/trntnsc.f90 and PW/cryst_to_car.f90. > > These subroutines are supposed to convert vectors > (cryst_to_car.f90) or matrices (trntnsc.f90) from crystal coordinates > to Cartesian (flag=1) or vice versa (flag=-1). And it seems like a > very straightforward task - just a little bit of vector algebra. But > when I derive these transformations myself they don't check out. Here > is what I've got. > > > > By definition of reciprocal vectors we have the following property for > dot-product: > > (a(i),b(j))= 2*pi*delta(i,j), (1a) > > where a(i) - lattice vectors, b(j) - reciprocal lattice vectors. In QE > these vectors are stored in matrices at(i,j) and bg(i,j) respectively. > Matrix elements of at are in units a_0 and matrix elements of bg are > in units (2*pi/a_0). So the condition (1a) can be rewritten as > follows: > > SUM(k){at(k,i)*bg(k,j)} = delta(i,j). (1b) > > In matrix form it can be represented as > > B' * A = I, or B' = inv(A), (2) > > where B' is a transposed matrix B and inv(A) is an inversed matrix A, > and I is the identity matrix. I checked these matrices for the case of > graphene and property (2) indeed holds true. > > > > Next, it's easy to show that transformation of an arbitrary vector > v from crystal to Cartesian coordinates is described by the matrix > equation: > > v_cart = A*v_cryst, (3a) I am not following all your notation, but it seems to me that you are missing the difference between vectors that in crystal coordinates are espressed in the basis of at, such as the position r, and vectors that in crystal coordinates are expressed in the basis of bg such as the k points for which Eq. 3a is not correct. HTH Andrea > > where A is a matrix at(i,j) mentioned above, which contains lattice > vector components. If you multiply equation (3a) from the left by > inv(A) and use the property inv(A)=B' (see eq. (2)) you will get the > reverse transformation: > > v_cryst = B' * v_cart (4a) > > > > Obviously, for an arbitrary matrix W the transformation equations will > be > > W_cart = A' *W_cryst * A (3b) > > W_cryst = B * W_cart * B' (4b) > > > > But that's not what I see in the code. In case of subroutine > trntnsc.f90 matrices A and B switch places in (3b) and (4b). And in > cryst_to_car.f90 everything seems fine but only if you use at(i,j) as > a transformation matrix when flag=1 and bg(i,j) if flag=-1. However, > when it's called in subroutine lint.f90 it looks like this: > > > > call cryst_to_cart (nkh,xp,at,-1) > > > > Like I said, I think the correct result can be obtained only in two > cases: > > call lint(nkh,xp,at,+1) - Crystal -> Cartesian > > call lint(nkh,xp,bg,-1) - Cartesian -> Crystal > > > > Maybe I missed something but this seems like a pretty simple problem > from vector algebra. Any ideas? > > > > > > Best regards, > > > > Kostyantyn Borysenko > > Department of Electrical and Computer Engineering > > North Carolina State University > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From sclauzer at sissa.it Fri May 22 12:06:26 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Fri, 22 May 2009 12:06:26 +0200 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back In-Reply-To: References: Message-ID: <4A167922.3040201@sissa.it> Kostyantyn Borysenko wrote: > Hi everyone, > > I'm having problems with the math behind some subroutines. I am talking > about PH/trntnsc.f90 and PW/cryst_to_car.f90. > > These subroutines are supposed to convert vectors (cryst_to_car.f90) or > matrices (trntnsc.f90) from crystal coordinates to Cartesian (flag=1) or > vice versa (flag=-1). And it seems like a very straightforward task > - just a little bit of vector algebra. But when I derive these > transformations myself they don't check out. Here is what I've got. > > > > By definition of reciprocal vectors we have the following property for > dot-product: > > (a(i),b(j))= 2*pi*delta(i,j), (1a) > > where a(i) - lattice vectors, b(j) - reciprocal lattice vectors. In QE > these vectors are stored in matrices at(i,j) and bg(i,j) respectively. > Matrix elements of at are in units a_0 and matrix elements of bg are in > units (2*pi/a_0). So the condition (1a) can be rewritten as follows: > > SUM(k){at(k,i)*bg(k,j)} = delta(i,j). (1b) > > In matrix form it can be represented as > > B' * A = I, or B' = inv(A), (2) > > where B' is a transposed matrix B and inv(A) is an inversed matrix A, > and I is the identity matrix. I checked these matrices for the case of > graphene and property (2) indeed holds true. > > > > Next, it's easy to show that transformation of an arbitrary vector > v from crystal to Cartesian coordinates is described by the matrix equation: > > v_cart = A*v_cryst, (3a) > > where A is a matrix at(i,j) mentioned above, which contains lattice > vector components. If you multiply equation (3a) from the left by inv(A) > and use the property inv(A)=B' (see eq. (2)) you will get the reverse > transformation: > > v_cryst = B' * v_cart (4a) > > > > Obviously, for an arbitrary matrix W the transformation equations will be > > W_cart = A' *W_cryst * A (3b) > > W_cryst = B * W_cart * B' (4b) Are you sure? If I understand correctly the meaning of "matrix in cartesian/crystal coordinates" (which for me means that v_cart'*W_cart*v_cart = v_crys'*W_crys*v_crys), then, following your notation v_cart = A*v_cryst so that v_cart' = v_cryst' * A' It follows v_cart' * W_cart * v_cart = v_crys' * A' * W_cart * A* v_crys Then W_crys = A' * W_cart * A I suppose that also (4b) is not correct... GS > > > > But that's not what I see in the code. In case of subroutine trntnsc.f90 > matrices A and B switch places in (3b) and (4b). And in cryst_to_car.f90 > everything seems fine but only if you use at(i,j) as a transformation > matrix when flag=1 and bg(i,j) if flag=-1. However, when it's called in > subroutine lint.f90 it looks like this: > > > > call cryst_to_cart (nkh,xp,at,-1) > > > > Like I said, I think the correct result can be obtained only in two cases: > > call lint(nkh,xp,at,+1) - Crystal -> Cartesian > > call lint(nkh,xp,bg,-1) - Cartesian -> Crystal > > > > Maybe I missed something but this seems like a pretty simple problem > from vector algebra. Any ideas? > > > > > > Best regards, > > > > Kostyantyn Borysenko > > Department of Electrical and Computer Engineering > > North Carolina State University > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From kboryse at ncsu.edu Fri May 22 20:02:38 2009 From: kboryse at ncsu.edu (Kostyantyn Borysenko) Date: Fri, 22 May 2009 14:02:38 -0400 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back Message-ID: <6E59C4FD0DDF4393A241193EC33E0EEC@ece.ncsu.edu> Dear Andrea and Gabriele, Thank you for your responses. 2Gabriele: I beleive there is a mistake in your first step. When you start with v_cart = A*v_cryst, you try to use the property (C*D)' = D' * C', which is true for the product of two matrices. In our case we have a product of matrix A and vector v_cryst. So you will get v_cart' = v_cryst' * A Also, in my original question I meant that vector v and matrix W represent some physical values that have correspongly vector and tensor nature. If one wants to change the coordinate system where these values are defined, one needs to perform an appropriate transformation. In our case this transformation is determined by the matrices A and B constructed from lattice vectors in real and reciprocal space. So if you want to transform a vector using transformation matrix T, you will get an equation of the form: v_new = T*v_old; and if you transform a matrix the corresponding equation will look like W_new = T' * W_old * T. But in the context of this discussion, the product of the form (v' * W * v) is irrelevant. I still can't see any false logic in my derivation. Best regards, Kostyantyn Borysenko Department of Electrical and Computer Engineering North Carolina State University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090522/b1b76809/attachment.htm From kboryse at ncsu.edu Fri May 22 22:40:39 2009 From: kboryse at ncsu.edu (Kostyantyn Borysenko) Date: Fri, 22 May 2009 16:40:39 -0400 Subject: [Pw_forum] Re: Coordinate transorms: Crystal to Cartesian and back (Dal Corso Andrea) Message-ID: <66D7C32163C24556BD5469F93B9248D7@ece.ncsu.edu> Dear Andrea, I finally figured it out, thanks to your hint. I guess I was misled by the term "crystal", which actually refers to the reciprocal lattice. Now everything checks out. Thanks again. Best regards, Kostyantyn Borysenko Department of Electrical and Computer Engineering North Carolina State University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090522/c17e907f/attachment.htm From pc229 at kent.ac.uk Sat May 23 16:47:23 2009 From: pc229 at kent.ac.uk (Pieremanuele Canepa) Date: Sat, 23 May 2009 15:47:23 +0100 Subject: [Pw_forum] DFT +U difficult compiling comp_resp_mat.j Message-ID: <8b443ba10905230747w79bee19p99e44870740b613e@mail.gmail.com> Dear all, I'm trying to compile the comp_resp_mat.j concerning the DFT+U. Since I am running the ifort on a 32 bit machine I replaced what was written in the original file as below: -lmkl_ia32 -lguide -lpthread ifort -o r.x resp_mat.f90 -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -lpthread What I got are these two lines /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_fun' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_dll' by looking at frevious message on PW_forum I found some else having my same problem. (link http://www.democritos.it/pipermail/pw_forum/2008-November/010596.html) So, What am I suppose to do ? anyone has already compiled this file under a 32 bit machine? Cheers, Piero -- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom ----------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090523/257e3ec0/attachment.htm From pc229 at kent.ac.uk Sat May 23 16:56:40 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Sat, 23 May 2009 15:56:40 +0100 Subject: [Pw_forum] DFT +U difficult compiling comp_resp_mat.j Message-ID: Dear all, I'm trying to compile the comp_resp_mat.j concerning the DFT+U. Since I am running the ifort on a 32 bit machine I replaced what was written in the original file as below: ifort -o r.x resp_mat.f90 -L/opt/intel/mkl/10.1.2.024/lib/32? -lmkl_lapack -lmkl_ia32 -lguide -lpthread What I got are these two lines? ?/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_fun' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_dll' by looking at the previous messages? on the PW_forum I found someone else having my same problem. (link http://www.democritos.it/pipermail/pw_forum/2008-November/010596.html) So, What am I suppose to do ?? anyone? has already compiled? this file under a 32 bit machine? Cheers, Piero --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090523/73e6d8fc/attachment.htm From pc229 at kent.ac.uk Sat May 23 16:55:16 2009 From: pc229 at kent.ac.uk (Pieremanuele Canepa) Date: Sat, 23 May 2009 15:55:16 +0100 Subject: [Pw_forum] DFT +U difficult compiling comp_resp_mat.j Message-ID: <8b443ba10905230755q271b253by9f256033f5b77324@mail.gmail.com> Dear all, I'm trying to compile the comp_resp_mat.j concerning the DFT+U. Since I am running the ifort on a 32 bit machine I replaced what was written in the original file as below: -lmkl_ia32 -lguide -lpthread ifort -o r.x resp_mat.f90 -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -lpthread What I got are these two lines /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_fun' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_dll' by looking at frevious message on PW_forum I found some else having my same problem. (link http://www.democritos.it/pipermail/pw_forum/2008-November/010596.html) So, What am I suppose to do ? anyone has already compiled this file under a 32 bit machine? Cheers, Piero -- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom ----------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090523/f4154cb7/attachment.htm From sophia.nishad at gmail.com Sat May 23 17:17:27 2009 From: sophia.nishad at gmail.com (Sophia Nishad) Date: Sat, 23 May 2009 10:17:27 -0500 Subject: [Pw_forum] Optical phonon convergence at Gamma-point Message-ID: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> Hi QE users, I am a new user of QE. So, wanted to run some test cases for Si. Energy calculations were ok, but phonons are not. I am having trouble to get good result of optical phonons for 2-atom Si. Convergence of energy is obtained with ecutwfc 25Ry k-mesh 6 6 6 0 0 0 norm-conserving PP conv_thr = 1.0d-12 tr2_ph=1.0d-20 Even though the energy difference was reduced w.r.t. increasing ecutwfc and k-mesh, I could not find good values for optical phonons ((I did not worry about ASR for this Gamma calculation)) w.r.t these quantities. Here's what I get with 6x6x6 k-mesh without any offset of k-mesh: 15Ry 519.587327 [cm-1] 16Ry 516.231236 [cm-1] 20Ry 508.216250 [cm-1] 30Ry 510.229073 [cm-1] 35Ry 510.044651 [cm-1] 50Ry 509.893556 [cm-1] 100Ry 511.548608 [cm-1] So some oscillations are there. But I would say it is converged around 510. The experimental value is ~519. So the QE result is quite different. I read the tutorials and archieve, but I did not find the answer of my problem. http://users.aims.ac.za/~paolo/tutorial_phon.htmat obtained the optical phonons to be 516.197992 [cm-1] for a ecutwfc of 16Ry. But at this ecutwfc the energy is not converged for Si. If I find ecutwfc to get energy convergence it becomes above 20Ry. But at this ecutwfc optical phonons are not getting better. I would expect better results at higher mesh and cutoff. I checked my result adding epsil=.true. as well. But result didn't change much. Could you suggest what might improve the result. Thanks in advance, -- Sophia Univ. of California - Berkeley, USA -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090523/dd6ef6d1/attachment.htm From yu.xie at mat.ethz.ch Sat May 23 21:01:57 2009 From: yu.xie at mat.ethz.ch (mat) Date: Sat, 23 May 2009 15:01:57 -0400 Subject: [Pw_forum] question about the format of gam.lines generated by matdyn.x Message-ID: <21ECEF25B0564508B283277F17322C39@yuxiePC> Dear all pwscf users: I tried to use QE for el-ph coupling calculations. I compiled espresso-4.0.5 on the cluster and the examples were all ok. When I was doing my own project, I got some problems with the format of 'gam.lines' file which generated by matdyn.x. The error message is like the following: PGFIO-F-235/formatted write/unit=20/edit descriptor does not match item type. File name = gam.lines formatted, sequential access record = 6 In source file matdyn.F90, at line number 1801 It seems like the description of gamma(:,: ) has some problems. Does anyone has some experiences with this problem? I'm very appreciated for your help! Have a nice weekend. Yu Xie -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090523/acf9a6d7/attachment.htm From panda.deng.pan at gmail.com Sat May 23 21:22:18 2009 From: panda.deng.pan at gmail.com (=?GB2312?B?xcu1xw==?=) Date: Sun, 24 May 2009 03:22:18 +0800 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> Message-ID: Dear Sophia, You didn't mention any value about ecutrho.I surpose that you have the default value (4 * ecutwfc ) which is enough for a good calculation. I can get more imformation from the directory '/Doc/input_pw' contained in QE file.The suggested value is about 8-10 times as the ecutwfc. Hope this might help! Cheers. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090524/0acfcfc2/attachment.htm From akohlmey at cmm.chem.upenn.edu Sat May 23 21:44:46 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 23 May 2009 15:44:46 -0400 Subject: [Pw_forum] DFT +U difficult compiling comp_resp_mat.j In-Reply-To: References: Message-ID: <1243107886.10549.6096.camel@zero> On Sat, 2009-05-23 at 15:56 +0100, pc229 at kent.ac.uk wrote: > Dear all, dear piero, > I'm trying to compile the comp_resp_mat.j concerning the DFT+U. Since > I am running the ifort on a 32 bit machine I replaced what was written > in the original file as below: > > ifort -o r.x resp_mat.f90 -L/opt/intel/mkl/10.1.2.024/lib/32 > -lmkl_lapack -lmkl_ia32 -lguide -lpthread > > What I got are these two lines > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > reference to `mkl_serv_load_fun' > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference > to `mkl_serv_load_dll' yes, this is the correct behavior. you cannot mix and match static and dynamic libraries with MKL. so, e.g. -lmkl_lapack -lmkl should work. our you'd have to use something more elaborate like explained in the MKL documentation. for example you normally want to link the serial MKL API kernel. cheers, axel. > by looking at the previous messages on the PW_forum I found someone > else having my same problem. (link > http://www.democritos.it/pipermail/pw_forum/2008-November/010596.html) > > So, What am I suppose to do ? anyone has already compiled this file > under a 32 bit machine? > Cheers, Piero > > --- > Pieremanuele Canepa > Room 230 > School of Physical Sciences, Ingram Building, > University of Kent, Canterbury, Kent, > CT2 7NH > United Kingdom > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From yu.xie at mat.ethz.ch Sat May 23 22:02:55 2009 From: yu.xie at mat.ethz.ch (mat) Date: Sat, 23 May 2009 16:02:55 -0400 Subject: [Pw_forum] question about the format of gam.lines generated by matdyn.x Message-ID: <7D5840C2573041209D755C1755C5782E@yuxiePC> Dear all, I found the reason. I recompiled the QE by using ifort and I got the gam.lines file. But there was another problem. The gam.lines file is looks like: Broadening 0.0050 1 0.0037 -0.0001 0.0059 0.0441 -0.0039 0.0409 0.1385 -0.0046 0.0098 ***** -0.1427 0.1015 -0.0148 0.2303 0.1364 -0.0261 0.3156 0.2340 0.9957 ***** 4.3933 2 0.0104 0.0000 0.0065 -0.0137 -0.0198 0.0594 0.1533 0.0025 0.0002 ***** -0.1494 0.1089 0.0046 0.2161 0.2436 -0.0276 0.4711 0.1582 1.0774 ***** 4.3485 As u can see, I have 21 modes and the 10th and 20th modes are *****. This is because in the line 1796 of matdyn.f90 'FORMAT( 3x,I5,' ',9F8.4 )'. Only ten data can be written into one line. The only solution I can do is change 9F8.4 to 21F8.4, but I'm not sure whether there will be another problem. Best regards, Yu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090523/d00b731b/attachment.htm From sophia.nishad at gmail.com Sun May 24 23:04:48 2009 From: sophia.nishad at gmail.com (Sophia Nishad) Date: Sun, 24 May 2009 14:04:48 -0700 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> Message-ID: <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> Hi, I have checked ecutrho as well (even though it is suggested in the mailing list that the default relation for ecutrho should be ok with norm-conserving pseudos). With changes to the ecutrho (~8,10 etc.) I see small change to the optical phonons' values (~1 cm-1), but it is still off from the experiemntal value by ~8cm-1. I was confused why no comment had been made in the link http://users.aims.ac.za/~paolo/tutorial_phon.htmat on the convergence of phonons with respect to ecutrho whereas the importance of the effects of ecutwfc on PW calculations was explained in so many places. Could I conclude that with the norm-conerving pseudo available on the QE website it is not possible to get improved phonon result??? It would be really helpful for me if someone could refer to results for Si that had been done checking the convergence of phonons w.r.t ecutwfc. Thanks, Sophia. On Sat, May 23, 2009 at 8:17 AM, Sophia Nishad wrote: > Hi QE users, > > I am a new user of QE. So, wanted to run some test cases for Si. Energy > calculations were ok, but phonons are not. I am having trouble to get good > result of optical phonons for 2-atom Si. Convergence of energy is obtained > with > > ecutwfc 25Ry > k-mesh 6 6 6 0 0 0 > norm-conserving PP > conv_thr = 1.0d-12 > tr2_ph=1.0d-20 > > Even though the energy difference was reduced w.r.t. increasing ecutwfc and > k-mesh, I could not find good values for optical phonons ((I did not worry > about ASR for this Gamma calculation)) w.r.t these quantities. Here's what I > get with 6x6x6 k-mesh without any offset of k-mesh: > > 15Ry 519.587327 [cm-1] > 16Ry 516.231236 [cm-1] > 20Ry 508.216250 [cm-1] > 30Ry 510.229073 [cm-1] > 35Ry 510.044651 [cm-1] > 50Ry 509.893556 [cm-1] > 100Ry 511.548608 [cm-1] > > So some oscillations are there. But I would say it is converged around 510. > The experimental value is ~519. So the QE result is quite different. > > I read the tutorials and archieve, but I did not find the answer of my > problem. > > http://users.aims.ac.za/~paolo/tutorial_phon.htmat > obtained the optical phonons to be 516.197992 [cm-1] for a ecutwfc of 16Ry. > But at this ecutwfc the energy is not converged for Si. If I find ecutwfc to > get energy convergence it becomes above 20Ry. But at this ecutwfc optical > phonons are not getting better. I would expect better results at higher mesh > and cutoff. I checked my result adding epsil=.true. as well. But result > didn't change much. > > Could you suggest what might improve the result. > > Thanks in advance, > -- > Sophia > Univ. of California - Berkeley, USA > -- Sophia -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090524/640710f9/attachment.htm From sclauzer at sissa.it Mon May 25 09:11:14 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 25 May 2009 09:11:14 +0200 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back In-Reply-To: <6E59C4FD0DDF4393A241193EC33E0EEC@ece.ncsu.edu> References: <6E59C4FD0DDF4393A241193EC33E0EEC@ece.ncsu.edu> Message-ID: <4A1A4492.2030804@sissa.it> Kostyantyn Borysenko wrote: > Dear Andrea and Gabriele, > > Thank you for your responses. > > 2Gabriele: I beleive there is a mistake in your first step. When you > start with v_cart = A*v_cryst, you try to use the property (C*D)' = D' * > C', which is true for the product of two matrices. In our case we have > a product of matrix A and vector v_cryst. So you will get v_cart' = > v_cryst' * A Dear Kostyantyn, maybe the linear algebra I'm speaking about is not the same you learnt... I believe mine works! GS > Also, in my original question I meant that vector v and matrix W > represent some physical values that have correspongly vector and tensor > nature. If one wants to change the coordinate system where these values > are defined, one needs to perform an appropriate transformation. In our > case this transformation is determined by the matrices A and B > constructed from lattice vectors in real and reciprocal space. So if you > want to transform a vector using transformation matrix T, you will get > an equation of the form: v_new = T*v_old; and if you transform a matrix > the corresponding equation will look like W_new = T' * W_old * T. > But > in the context of this discussion, the product of the form (v' * W * v) > is irrelevant. > I still can't see any false logic in my derivation. > > > Best regards, > > > > Kostyantyn Borysenko > > Department of Electrical and Computer Engineering > North Carolina State University > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From kazempoor2000 at yahoo.com Mon May 25 09:48:14 2009 From: kazempoor2000 at yahoo.com (ali kazempoor) Date: Mon, 25 May 2009 00:48:14 -0700 (PDT) Subject: [Pw_forum] temperature Message-ID: <215224.61350.qm@web112505.mail.gq1.yahoo.com> Dear all ?I want to study defect formation of? charged defect at required temperature . Since DFT is at T=0 , So I think that when we introduce pressure on system it is equal to studying system at some finite temperature. Is it true? and what is the relation and formula that shows how can we obtain the temperature by knowing the pressure of the system? thanks a lot Ali Kazempour Physics department, Isfahan University of Technology 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/9a08f2e9/attachment.htm From giannozz at democritos.it Mon May 25 10:26:50 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 25 May 2009 10:26:50 +0200 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> Message-ID: <5382CB05-00DC-462B-8F1A-91D299697528@democritos.it> On May 24, 2009, at 23:04 , Sophia Nishad wrote: > it is still off from the experiemntal value by ~8cm-1. both pseudopotentials and approximations to the exchange-correlation functional introduce intrinsic errors. It is quite hard to get any better results than those you get. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From anvari_meh at physics.iust.ac.ir Mon May 25 12:24:14 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Mon, 25 May 2009 13:54:14 +0330 Subject: [Pw_forum] van der waals force Message-ID: Thanks a lot because of your attention. I am trying to do phonon calculation in graphene so I think, having well approximation for pseudo potential & exchange-correlation one, will help you to achieve good frequencies which are compatible with experiment. Now I think in VdW force you should have segnificient changes in your core density, because of polarization. It seems that situation could be same when we use full potential (Isn't true?) Best Regards Mehrnaz Anvari Iran University Of Science & Technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/df298ce6/attachment.htm From paulatto at sissa.it Mon May 25 12:30:03 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 25 May 2009 12:30:03 +0200 Subject: [Pw_forum] van der waals force In-Reply-To: References: Message-ID: In data 25 maggio 2009 alle ore 12:24:14, Mehrnaz Anvari ha scritto: > Thanks a lot because of your attention. I > am trying to do phonon calculation in graphene so I think, having > well approximation for pseudo potential & exchange-correlation > one, will help you to achieve good frequencies which are compatible with > experiment. If I new an exchange-correlation functional guaranteed to reproduce experimental values I would be the younger professor of condensed matter physics... unluckily all available XC functionals involve some level of aproximation and can only reproduce experiments up to that level. There is a lot of literature on this topic; I suggest consulting the book by prof. Richard Martin. > Now I think in VdW force you should have segnificient changes in your > core density, because of polarization. I really don't know, VdW interaction is assumend to be relatively weak in the scale of electronic interaction. Hence they should not affect the ionic cores much. Again, an uncountable number of papers have been written on this topic, sometimes VdW forces can be neglected, sometimes they are the only term keeping the system together. > It seems that situation could be same when we use full potential (Isn't > true?) You can use PAW instead of ultrasoft pseudopotential to improve on the core aproximation. Keep in mind that pseudopotential and frozen-core are two different kinds of aproximation: the ultrasoft/norm-conserving methods use them both, while PAW only uses frozen-core. In general, this has nothing to do with VdW forces. If your concern is the lack of VdW interactions, I don't see how improving the threatment of core orbitals can help (but I may be mistaken). Furthermore, QE does not implement any empiric or equally cheap VdW correction (left alone doing phonon calculations with it), hence if you need it you have to look somewhere else, or implement it yourself. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From pc229 at kent.ac.uk Mon May 25 13:14:36 2009 From: pc229 at kent.ac.uk (Pieremanuele Canepa) Date: Mon, 25 May 2009 12:14:36 +0100 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U Message-ID: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> Dear all, I'm trying to compile the comp_resp_mat.j concerning the DFT+U. Then, I tried to use the default settings (ifort-o r.x resp_mat.f90 -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 -lguide -lpthread) and what I got, is reported below. Do you have any suggestions?Probably I am making some stupid mistake. have somebody already compiled this program on 32 bit machine? I don't have any experience compiling with lapack libraries as you can probably realize. Cheers, Piero resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. /tmp/ifortMUANg9.o: In function `invmat_.': resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_' resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_' resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_pds_c_blklu_unsym_omp_pardiso' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_lapack_cpptrs' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_strcat2' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_pds_blklu_omp_pardiso' and so on -- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom ----------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/e250b45c/attachment.htm From sophia.nishad at gmail.com Mon May 25 15:21:37 2009 From: sophia.nishad at gmail.com (Sophia Nishad) Date: Mon, 25 May 2009 06:21:37 -0700 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <5382CB05-00DC-462B-8F1A-91D299697528@democritos.it> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> <5382CB05-00DC-462B-8F1A-91D299697528@democritos.it> Message-ID: <8a0137e60905250621v3933422obc885a644e5dfbb0@mail.gmail.com> Thanks, Paolo. I have little confusion about your comment though. Earlier I had used norm-conserving PP in supercell calculations as well; that produced converged frequency of 518.2 for Si. So, is it possible that linear response calculations might not give better results compared to supercell calculations (which is very expensive). The result could be sensitive to the assumptions of linear response theory as well?? Thanks, Sophia On Mon, May 25, 2009 at 1:26 AM, Paolo Giannozzi wrote: > > On May 24, 2009, at 23:04 , Sophia Nishad wrote: > > > it is still off from the experiemntal value by ~8cm-1. > > both pseudopotentials and approximations to the exchange-correlation > functional introduce intrinsic errors. It is quite hard to get any > better > results than those you get. > > Paolo > --- > Paolo Giannozzi, Democritos and University of Udine, Italy > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Sophia -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/d21eef9d/attachment-0001.htm From baroni at sissa.it Mon May 25 15:59:47 2009 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 25 May 2009 15:59:47 +0200 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> Message-ID: On May 24, 2009, at 11:04 PM, Sophia Nishad wrote: > Hi, Sohia: > I have checked ecutrho as well (even though it is suggested in the > mailing list that the default relation for ecutrho should be ok with > norm-conserving pseudos). in fact, with NC PPs there is no point in increasing the charge cutoff beyon 4 x wvfct cutoff > With changes to the ecutrho (~8,10 etc.) I see small change to the > optical phonons' values (~1 cm-1), but it is still off from the > experiemntal value by ~8cm-1. 8/510 = 1.5% I highly value your expectations abouT DFT, linear response, QE, and all the rest, still 1.5% seems to me an excellent agreement with experiments > I was confused why no comment had been made in the link http://users.aims.ac.za/~paolo/tutorial_phon.htmat > on the convergence of phonons with respect to ecutrho whereas the > importance of the effects of ecutwfc on PW calculations was > explained in so many places. > > Could I conclude that with the norm-conerving pseudo available on > the QE website it is not possible to get improved phonon result??? > > It would be really helpful for me if someone could refer to results > for Si that had been done checking the convergence of phonons w.r.t > ecutwfc. well done linear-response phonon calculations should just reproduce frozen-phonon calculations, to which all the explanations about total- energy calculations apply. there are residual subtelties regarding the fulfillment of the acoustic sum rule, but these mainly affect acoustic phonons. I will explain the little I have understood on this issue in another occasion. SB > --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/a679d273/attachment.htm From sclauzer at sissa.it Mon May 25 16:18:02 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 25 May 2009 16:18:02 +0200 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> Message-ID: <4A1AA89A.8010701@sissa.it> I thought Axel already replied to this post... Pieremanuele Canepa wrote: > Dear all, > I'm trying to compile the comp_resp_mat.j concerning the DFT+U. Then, I > tried to use the default settings (ifort-o r.x resp_mat.f90 > -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 -lguide > -lpthread) and what I got, is reported below. Do you have any suggestions? > Probably I am making some stupid mistake. have somebody already > compiled this program on 32 bit machine? > I don't have any experience compiling with lapack libraries as you can > probably realize. > Cheers, Piero I don't understand if you're trying to link against mkl 7.2 or 10.1. Do you know at least which version is installed on your machine and where? Try to find out this first. On my 32 bit Intel machine I've got MKL 9.0 installed in /usr/local/intel_mkl_90/ and this compilation command works (for static linking): ifort -o resp_mat.x resp_mat.f90 \ -L"/usr/local/intel_mkl_90/lib/32" -lmkl_lapack -lmkl_ia32 -lguide -lpthread ldd resp_mat.x linux-gate.so.1 => (0x00af3000) libguide.so => /usr/local/intel_mkl_90/lib/32/libguide.so (0x00a55000) libm.so.6 => /lib/libm.so.6 (0x00aa4000) libc.so.6 => /lib/libc.so.6 (0x00111000) libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x00d4c000) libdl.so.2 => /lib/libdl.so.2 (0x00acd000) libpthread.so.0 => /lib/libpthread.so.0 (0x00ad3000) /lib/ld-linux.so.2 (0x00941000) If you want to link dynamically use: ifort -o resp_mat.x resp_mat.f90 \ -L"/usr/local/intel_mkl_90/lib/32" -lmkl_lapack64 -lmkl -lguide -lpthread ldd resp_mat.x linux-gate.so.1 => (0x00741000) libmkl_lapack64.so => /usr/local/intel_mkl_90/lib/32/libmkl_lapack64.so (0x00c32000) libmkl.so => /usr/local/intel_mkl_90/lib/32/libmkl.so (0x00eba000) libguide.so => /usr/local/intel_mkl_90/lib/32/libguide.so (0x00111000) libpthread.so.0 => /lib/libpthread.so.0 (0x00ad3000) libm.so.6 => /lib/libm.so.6 (0x00aa4000) libc.so.6 => /lib/libc.so.6 (0x0095f000) libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x0015e000) libdl.so.2 => /lib/libdl.so.2 (0x00acd000) /lib/ld-linux.so.2 (0x00941000) Obviously you must change the path to the exact location of your MKL installation HTH GS > > resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > /tmp/ifortMUANg9.o: In function `invmat_.': > resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_' > resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_' > resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_' > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > : undefined reference to > `mkl_pds_c_blklu_unsym_omp_pardiso' > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > : undefined reference to > `mkl_lapack_cpptrs' > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > : undefined reference to > `mkl_serv_strcat2' > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > : undefined reference to > `mkl_pds_blklu_omp_pardiso' > and so on > -- > Pieremanuele Canepa > Room 230 > School of Physical Sciences, Ingram Building, > University of Kent, Canterbury, Kent, > CT2 7NH > United Kingdom > ----------------------------------------------------------- > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From pc229 at kent.ac.uk Mon May 25 16:48:54 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Mon, 25 May 2009 15:48:54 +0100 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: <4A1AA89A.8010701@sissa.it> References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> Message-ID: I do apologize, I made a mistake? when I paste the string that I use to compile it. Anyway I tried both the static and the dynamics compilation using either this string and the output is : and ifort?-o resp_mat.x resp_mat.f90 \ -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack64 -lmkl -lguide -lpthread resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. ld: cannot find -lmkl_lapack64 I cannot manage this problem. If it doesn't matter, can you please send me, privately, your compiled version? I am struggling a lot. Thank you, Piero ----- Original Message ----- From: Gabriele Sclauzero Date: Monday, May 25, 2009 3:18 pm Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U To: PWSCF Forum > I thought Axel already replied to this post... > > Pieremanuele Canepa wrote: > > Dear all, > > I'm trying to compile the comp_resp_mat.j concerning the > DFT+U. Then, I > > tried to use the default settings (ifort-o r.x resp_mat.f90 > > -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 - > lguide > > -lpthread) and what I got, is reported below. Do you have any > suggestions?> Probably? I am making some stupid mistake. > have somebody already > > compiled this program on 32 bit machine? > > I don't have any experience compiling with lapack libraries as > you can > > probably realize. > > Cheers, Piero > > I don't understand if you're trying to link against mkl 7.2 or > 10.1. Do you know at least > which version is installed on your machine and where? Try to > find out this first. > > On my 32 bit Intel machine I've got MKL 9.0 installed in > /usr/local/intel_mkl_90/ and this > compilation command works (for static linking): > > ifort -o resp_mat.x resp_mat.f90 \ > ??? -L"/usr/local/intel_mkl_90/lib/32" - > lmkl_lapack -lmkl_ia32 -lguide -lpthread > > ldd resp_mat.x > ???????? linux-gate.so.1 > =>? (0x00af3000) > ???????? libguide.so => > /usr/local/intel_mkl_90/lib/32/libguide.so (0x00a55000) > ???????? libm.so.6 => > /lib/libm.so.6 (0x00aa4000) > ???????? libc.so.6 => > /lib/libc.so.6 (0x00111000) > ???????? libgcc_s.so.1 > => /lib/libgcc_s.so.1 (0x00d4c000) > ???????? libdl.so.2 => > /lib/libdl.so.2 (0x00acd000) > ???????? libpthread.so.0 > => /lib/libpthread.so.0 (0x00ad3000) > ???????? /lib/ld- > linux.so.2 (0x00941000) > > > If you want to link dynamically use: > > ifort -o resp_mat.x resp_mat.f90 \ > ??? -L"/usr/local/intel_mkl_90/lib/32" - > lmkl_lapack64 -lmkl -lguide -lpthread > > ldd resp_mat.x > ???????? linux-gate.so.1 > =>? (0x00741000) > ???????? > libmkl_lapack64.so => > /usr/local/intel_mkl_90/lib/32/libmkl_lapack64.so (0x00c32000) > ???????? libmkl.so => > /usr/local/intel_mkl_90/lib/32/libmkl.so (0x00eba000) > ???????? libguide.so => > /usr/local/intel_mkl_90/lib/32/libguide.so (0x00111000) > ???????? libpthread.so.0 > => /lib/libpthread.so.0 (0x00ad3000) > ???????? libm.so.6 => > /lib/libm.so.6 (0x00aa4000) > ???????? libc.so.6 => > /lib/libc.so.6 (0x0095f000) > ???????? libgcc_s.so.1 > => /lib/libgcc_s.so.1 (0x0015e000) > ???????? libdl.so.2 => > /lib/libdl.so.2 (0x00acd000) > ???????? /lib/ld- > linux.so.2 (0x00941000) > > Obviously you must change the path to the exact location of your > MKL installation > > > HTH > > GS > > > > > > resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > > resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > > resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > > resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > > /tmp/ifortMUANg9.o: In function `invmat_.': > > resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_' > > resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_' > > resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_pds_c_blklu_unsym_omp_pardiso' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_lapack_cpptrs' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_serv_strcat2' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_pds_blklu_omp_pardiso' > > and so on > > -- > > Pieremanuele Canepa > > Room 230 > > School of Physical Sciences, Ingram Building, > > University of Kent, Canterbury, Kent, > > CT2 7NH > > United Kingdom > > ----------------------------------------------------------- > > > > > > --------------------------------------------------------------- > --------- > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD > Student????????????????? | > | c/o:?? SISSA & CNR-INFM > Democritos,????????????? | > |??????? via Beirut 2-4, > 34014 Trieste (Italy)???? | > | email: > sclauzer at sissa.it???????????????????????? | > | phone: +39 040 3787 > 511????????????????????????? | > | skype: > gurlonotturno???????????????????????????? | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/858bbed7/attachment-0001.htm From pc229 at kent.ac.uk Mon May 25 16:51:14 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Mon, 25 May 2009 15:51:14 +0100 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: <4A1AA89A.8010701@sissa.it> References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> Message-ID: I do apologize, I made a mistake? when I paste the string that I use to compile it. Anyway I tried both the static and the dynamics compilation using either this string ifort -o resp_mat.x resp_mat.f90 \ -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -lpthread and the output is : resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_fun' /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_dll' and ifort?-o resp_mat.x resp_mat.f90 \ -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack64 -lmkl -lguide -lpthread resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. ld: cannot find -lmkl_lapack64 I cannot manage this problem. If it doesn't matter, can you please send me, privately, your compiled version? I am struggling a lot. Thank you, Piero ----- Original Message ----- From: Gabriele Sclauzero Date: Monday, May 25, 2009 3:18 pm Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U To: PWSCF Forum > I thought Axel already replied to this post... > > Pieremanuele Canepa wrote: > > Dear all, > > I'm trying to compile the comp_resp_mat.j concerning the > DFT+U. Then, I > > tried to use the default settings (ifort-o r.x resp_mat.f90 > > -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 - > lguide > > -lpthread) and what I got, is reported below. Do you have any > suggestions?> Probably? I am making some stupid mistake. > have somebody already > > compiled this program on 32 bit machine? > > I don't have any experience compiling with lapack libraries as > you can > > probably realize. > > Cheers, Piero > > I don't understand if you're trying to link against mkl 7.2 or > 10.1. Do you know at least > which version is installed on your machine and where? Try to > find out this first. > > On my 32 bit Intel machine I've got MKL 9.0 installed in > /usr/local/intel_mkl_90/ and this > compilation command works (for static linking): > > ifort -o resp_mat.x resp_mat.f90 \ > ??? -L"/usr/local/intel_mkl_90/lib/32" - > lmkl_lapack -lmkl_ia32 -lguide -lpthread > > ldd resp_mat.x > ???????? linux-gate.so.1 > =>? (0x00af3000) > ???????? libguide.so => > /usr/local/intel_mkl_90/lib/32/libguide.so (0x00a55000) > ???????? libm.so.6 => > /lib/libm.so.6 (0x00aa4000) > ???????? libc.so.6 => > /lib/libc.so.6 (0x00111000) > ???????? libgcc_s.so.1 > => /lib/libgcc_s.so.1 (0x00d4c000) > ???????? libdl.so.2 => > /lib/libdl.so.2 (0x00acd000) > ???????? libpthread.so.0 > => /lib/libpthread.so.0 (0x00ad3000) > ???????? /lib/ld- > linux.so.2 (0x00941000) > > > If you want to link dynamically use: > > ifort -o resp_mat.x resp_mat.f90 \ > ??? -L"/usr/local/intel_mkl_90/lib/32" - > lmkl_lapack64 -lmkl -lguide -lpthread > > ldd resp_mat.x > ???????? linux-gate.so.1 > =>? (0x00741000) > ???????? > libmkl_lapack64.so => > /usr/local/intel_mkl_90/lib/32/libmkl_lapack64.so (0x00c32000) > ???????? libmkl.so => > /usr/local/intel_mkl_90/lib/32/libmkl.so (0x00eba000) > ???????? libguide.so => > /usr/local/intel_mkl_90/lib/32/libguide.so (0x00111000) > ???????? libpthread.so.0 > => /lib/libpthread.so.0 (0x00ad3000) > ???????? libm.so.6 => > /lib/libm.so.6 (0x00aa4000) > ???????? libc.so.6 => > /lib/libc.so.6 (0x0095f000) > ???????? libgcc_s.so.1 > => /lib/libgcc_s.so.1 (0x0015e000) > ???????? libdl.so.2 => > /lib/libdl.so.2 (0x00acd000) > ???????? /lib/ld- > linux.so.2 (0x00941000) > > Obviously you must change the path to the exact location of your > MKL installation > > > HTH > > GS > > > > > > resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > > resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > > resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > > resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > > /tmp/ifortMUANg9.o: In function `invmat_.': > > resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_' > > resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_' > > resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_pds_c_blklu_unsym_omp_pardiso' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_lapack_cpptrs' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_serv_strcat2' > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > : undefined > reference to > > `mkl_pds_blklu_omp_pardiso' > > and so on > > -- > > Pieremanuele Canepa > > Room 230 > > School of Physical Sciences, Ingram Building, > > University of Kent, Canterbury, Kent, > > CT2 7NH > > United Kingdom > > ----------------------------------------------------------- > > > > > > --------------------------------------------------------------- > --------- > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD > Student????????????????? | > | c/o:?? SISSA & CNR-INFM > Democritos,????????????? | > |??????? via Beirut 2-4, > 34014 Trieste (Italy)???? | > | email: > sclauzer at sissa.it???????????????????????? | > | phone: +39 040 3787 > 511????????????????????????? | > | skype: > gurlonotturno???????????????????????????? | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/0488074e/attachment-0001.htm From houdong_phy at mail.sdu.edu.cn Mon May 25 17:05:48 2009 From: houdong_phy at mail.sdu.edu.cn (Hou Dong) Date: Mon, 25 May 2009 23:05:48 +0800 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: <443262146.09322@mail.sdu.edu.cn> References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com><4A1AA89A.8010701@sissa.it> <443262146.09322@mail.sdu.edu.cn> Message-ID: <443262984.09323@mail.sdu.edu.cn> From: pc229 at kent.ac.uk Sent: Monday, May 25, 2009 10:51 PM To: PWSCF Forum Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U >Anyway I tried both the static and the dynamics compilation using either this string >ifort -o resp_mat.x resp_mat.f90 \ >-L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -lpthread try -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lmkl_core -lguide -lpthread >and the output is : .... >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_fun' >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_dll' it is related to a new feature of Intel MKL version 10 and higher. add -lmkl_core may solve these problems. ------------------------------------ Dong Hou School of Physics , Shandong University Jinan, Shandong ,P.R. China. houdong_phy at mail.sdu.edu.cn --------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/58d6d237/attachment.htm From pc229 at kent.ac.uk Mon May 25 17:20:57 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Mon, 25 May 2009 16:20:57 +0100 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: <443262984.09323@mail.sdu.edu.cn> References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> <443262146.09322@mail.sdu.edu.cn> <443262984.09323@mail.sdu.edu.cn> Message-ID: Thank you very much Dong Hou, now I can compile it correctly! ----- Original Message ----- From: Hou Dong Date: Monday, May 25, 2009 4:11 pm Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U To: PWSCF Forum > > > > From: pc229 at kent.ac.uk > Sent: Monday, May 25, 2009 10:51 PM > To: PWSCF Forum > Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j > to optimize the U > > >Anyway I tried both the static and the dynamics compilation > using either this string > >ifort -o resp_mat.x resp_mat.f90 \ > >-L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 - > lguide -lpthread > > ???? try -L/opt/intel/mkl/10.1.2.024/lib/32 - > lmkl_lapack -lmkl_ia32 -lmkl_core -lguide -lpthread > > >and the output is : > .... > >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > reference to `mkl_serv_load_fun' > >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > reference to `mkl_serv_load_dll' > > ??? it is related to a new feature of Intel MKL > version 10 and higher. add -lmkl_core may solve these problems. > > > ------------------------------------ > Dong Hou > School of Physics , Shandong University > Jinan, Shandong ,P.R. China. > houdong_phy at mail.sdu.edu.cn > --------------------------------------------- > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/b63b5b0d/attachment.htm From akohlmey at cmm.chem.upenn.edu Mon May 25 17:22:15 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 May 2009 11:22:15 -0400 Subject: [Pw_forum] van der waals force In-Reply-To: References: Message-ID: <1243264935.10549.8970.camel@zero> On Mon, 2009-05-25 at 13:54 +0330, Mehrnaz Anvari wrote: > Thanks a lot because of your attention. I am trying to do phonon > calculation in graphene so I think, having well approximation for > pseudo potential & exchange-correlation one, will help you to achieve > good frequencies which are compatible with experiment. > Now I think in VdW force you should have segnificient changes in your > core density, because of polarization. It seems that situation could why so? why should the core electrons be more polarizable? i'd say on the contrary. please note that dispersion interactions are interactions between induced dipoles, thus quite weak compared to point charge interactions. now, the closer you get to the core, the less the core charge is screened by the electrons and thus the changes due to polarization would matter increasingly less the closer you get to the core. this is why van der waals interactions matter the most for large and polarizable elements and diffuse electron densities. > be same when we use full potential (Isn't true?) no. as nicola pointed out, it can matter how many electrons you put into the valence depending on the element, but the (inner) core electrons for sure do not matter. keep in mind that people successfully study much more radical polarizations than what dispersion does with pseudopotentials. if you were right, all calculations with pseudopotentials would be useless. cheers, axel. > Best Regards > Mehrnaz Anvari > Iran University Of Science & Technology > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From pc229 at kent.ac.uk Mon May 25 17:32:55 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Mon, 25 May 2009 16:32:55 +0100 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> <443262146.09322@mail.sdu.edu.cn> <443262984.09323@mail.sdu.edu.cn> Message-ID: Actually, the compilation is fine goes fine since I try to run the executable. The outcome is: resp_mat.x: error while loading shared libraries: libmkl_lapack.so: cannot open shared object file: No such file or directory. Has anyone figured out how to do ? Thank you ----- Original Message ----- From: pc229 at kent.ac.uk Date: Monday, May 25, 2009 4:21 pm Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U To: PWSCF Forum > Thank you very much Dong Hou, now I can compile it correctly! > > ----- Original Message ----- > From: Hou Dong > Date: Monday, May 25, 2009 4:11 pm > Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j > to optimize the U > To: PWSCF Forum > > > > > > > > > From: pc229 at kent.ac.uk > > Sent: Monday, May 25, 2009 10:51 PM > > To: PWSCF Forum > > Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j > > to optimize the U > > > > >Anyway I tried both the static and the dynamics compilation > > using either this string > > >ifort -o resp_mat.x resp_mat.f90 \ > > >-L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 - > > lguide -lpthread > > > > ???? try -L/opt/intel/mkl/10.1.2.024/lib/32 - > > lmkl_lapack -lmkl_ia32 -lmkl_core -lguide -lpthread > > > > >and the output is : > > .... > > >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > > reference to `mkl_serv_load_fun' > > >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > > reference to `mkl_serv_load_dll' > > > > ??? it is related to a new feature of Intel MKL > > version 10 and higher. add -lmkl_core may solve these problems. > > > > > > ------------------------------------ > > Dong Hou > > School of Physics , Shandong University > > Jinan, Shandong ,P.R. China. > > houdong_phy at mail.sdu.edu.cn > > --------------------------------------------- > > > > --- > Pieremanuele Canepa > Room 230 > School of Physical Sciences, Ingram Building, > University of Kent, Canterbury, Kent, > CT2 7NH > United Kingdom > > > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/1168c3ba/attachment.htm From giannozz at democritos.it Mon May 25 17:39:37 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Mon, 25 May 2009 17:39:37 +0200 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> Message-ID: ifort -i-static -openmp -o resp_mat.x resp_mat.f90 \ -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_ia32 (assuming you have recent and properly installed ifort and mkl) P. From akohlmey at cmm.chem.upenn.edu Mon May 25 17:41:05 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 May 2009 11:41:05 -0400 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <8a0137e60905250621v3933422obc885a644e5dfbb0@mail.gmail.com> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> <5382CB05-00DC-462B-8F1A-91D299697528@democritos.it> <8a0137e60905250621v3933422obc885a644e5dfbb0@mail.gmail.com> Message-ID: <1243266065.10549.9013.camel@zero> On Mon, 2009-05-25 at 06:21 -0700, Sophia Nishad wrote: > > Thanks, Paolo. I have little confusion about your comment though. > > Earlier I had used norm-conserving PP in supercell calculations as > well; that produced converged frequency of 518.2 for Si. So, is it > possible that linear response calculations might not give better > results compared to supercell calculations (which is very expensive). > The result could be sensitive to the assumptions of linear response > theory as well?? no. there also is the error induced by using the electron density and the potential on a finite grid. those are somewhat random, depending on choice of k-points and atom positions and availability of FFT grids. if you get less than 2% deviation from experiment then this is rather by chance than by constructions. cheers, axel. > > Thanks, > Sophia > > > On Mon, May 25, 2009 at 1:26 AM, Paolo Giannozzi > wrote: > > On May 24, 2009, at 23:04 , Sophia Nishad wrote: > > > it is still off from the experiemntal value by ~8cm-1. > > > both pseudopotentials and approximations to the > exchange-correlation > functional introduce intrinsic errors. It is quite hard to get > any > better > results than those you get. > > Paolo > --- > Paolo Giannozzi, Democritos and University of Udine, Italy > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > Sophia > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From paulatto at sissa.it Mon May 25 17:43:01 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Mon, 25 May 2009 17:43:01 +0200 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> <443262146.09322@mail.sdu.edu.cn> <443262984.09323@mail.sdu.edu.cn> Message-ID: In data 25 maggio 2009 alle ore 17:32:55, ha scritto: > resp_mat.x: error while loading shared libraries: libmkl_lapack.so: > cannot open shared object file: No such file or directory. > Has anyone figured out how to do ? In the mkl installation directory there are some script, depending on the version they may have different name, but it may be something like mklvars32.sh ("find /dir/of/mkl -name \*.sh" usually finds it) Executing the script sets all the environment variables to the value required to run your application; to make the changes permanent you can copy the script to /etc/profile.d or call (as in ". /path/to/script.sh") it from your ~/.profile file. A similar script exist for ifort, please spend some time to check if your environment is configured correctly. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/ From akohlmey at cmm.chem.upenn.edu Mon May 25 17:51:55 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 May 2009 11:51:55 -0400 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> Message-ID: <1243266715.10549.9045.camel@zero> On Mon, 2009-05-25 at 15:48 +0100, pc229 at kent.ac.uk wrote: > I do apologize, > I made a mistake when I paste the string that I use to compile it. > Anyway I tried both the static and the dynamics compilation using > either this string > > and the output is : > > > and > ifort -o resp_mat.x resp_mat.f90 \ > -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack64 -lmkl -lguide > -lpthread why suddenly use now -lmkl_lapack64 ??? there is not libmkl_lapack64.so in the 32bit version of mkl. it should work with plain -lmkl_lapack cheers, axel. > I cannot manage this problem. If it doesn't matter, can you please > send me, privately, your compiled version? I am struggling a lot. > Thank you, Piero _please_ rather then doing random stuff that does not make sense, spend some time reading the MKL documentation. it is quite detailed, useful and explains _everything_ about linking MKL dynamically and statically. you have to keep in mind that the linux linker always prefers shared libraries (.so) over static libraries. you should also see from the documentation that the static version or lapack is already included in libmkl_ia32.a. in conclusion: either link with -openmp -lmkl_lapack -lmkl or -i-static -openmp -lmkl_ia32 or -lmkl_lapack -lmkl_intel -lmkl_sequential -lmkl_core cheers, axel. > ----- Original Message ----- > From: Gabriele Sclauzero > Date: Monday, May 25, 2009 3:18 pm > Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to > optimize the U > To: PWSCF Forum > > > I thought Axel already replied to this post... > > > > Pieremanuele Canepa wrote: > > > Dear all, > > > I'm trying to compile the comp_resp_mat.j concerning the > > DFT+U. Then, I > > > tried to use the default settings (ifort-o r.x resp_mat.f90 > > > -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 - > > lguide > > > -lpthread) and what I got, is reported below. Do you have any > > suggestions?> Probably I am making some stupid mistake. > > have somebody already > > > compiled this program on 32 bit machine? > > > I don't have any experience compiling with lapack libraries as > > you can > > > probably realize. > > > Cheers, Piero > > > > I don't understand if you're trying to link against mkl 7.2 or > > 10.1. Do you know at least > > which version is installed on your machine and where? Try to > > find out this first. > > > > On my 32 bit Intel machine I've got MKL 9.0 installed in > > /usr/local/intel_mkl_90/ and this > > compilation command works (for static linking): > > > > ifort -o resp_mat.x resp_mat.f90 \ > > -L"/usr/local/intel_mkl_90/lib/32" - > > lmkl_lapack -lmkl_ia32 -lguide -lpthread > > > > ldd resp_mat.x > > linux-gate.so.1 > > => (0x00af3000) > > libguide.so => > > /usr/local/intel_mkl_90/lib/32/libguide.so (0x00a55000) > > libm.so.6 => > > /lib/libm.so.6 (0x00aa4000) > > libc.so.6 => > > /lib/libc.so.6 (0x00111000) > > libgcc_s.so.1 > > => /lib/libgcc_s.so.1 (0x00d4c000) > > libdl.so.2 => > > /lib/libdl.so.2 (0x00acd000) > > libpthread.so.0 > > => /lib/libpthread.so.0 (0x00ad3000) > > /lib/ld- > > linux.so.2 (0x00941000) > > > > > > If you want to link dynamically use: > > > > ifort -o resp_mat.x resp_mat.f90 \ > > -L"/usr/local/intel_mkl_90/lib/32" - > > lmkl_lapack64 -lmkl -lguide -lpthread > > > > ldd resp_mat.x > > linux-gate.so.1 > > => (0x00741000) > > > > libmkl_lapack64.so => > > /usr/local/intel_mkl_90/lib/32/libmkl_lapack64.so (0x00c32000) > > libmkl.so => > > /usr/local/intel_mkl_90/lib/32/libmkl.so (0x00eba000) > > libguide.so => > > /usr/local/intel_mkl_90/lib/32/libguide.so (0x00111000) > > libpthread.so.0 > > => /lib/libpthread.so.0 (0x00ad3000) > > libm.so.6 => > > /lib/libm.so.6 (0x00aa4000) > > libc.so.6 => > > /lib/libc.so.6 (0x0095f000) > > libgcc_s.so.1 > > => /lib/libgcc_s.so.1 (0x0015e000) > > libdl.so.2 => > > /lib/libdl.so.2 (0x00acd000) > > /lib/ld- > > linux.so.2 (0x00941000) > > > > Obviously you must change the path to the exact location of your > > MKL installation > > > > > > HTH > > > > GS > > > > > > > > > > resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > > > resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > > > resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. > > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. > > > resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. > > > resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. > > > /tmp/ifortMUANg9.o: In function `invmat_.': > > > resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_' > > > resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_' > > > resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_' > > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > > : undefined > > reference to > > > `mkl_pds_c_blklu_unsym_omp_pardiso' > > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > > : undefined > > reference to > > > `mkl_lapack_cpptrs' > > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > > : undefined > > reference to > > > `mkl_serv_strcat2' > > > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so > > > : undefined > > reference to > > > `mkl_pds_blklu_omp_pardiso' > > > and so on > > > -- > > > Pieremanuele Canepa > > > Room 230 > > > School of Physical Sciences, Ingram Building, > > > University of Kent, Canterbury, Kent, > > > CT2 7NH > > > United Kingdom > > > ----------------------------------------------------------- > > > > > > > > > --------------------------------------------------------------- > > --------- > > > > > > _______________________________________________ > > > Pw_forum mailing list > > > Pw_forum at pwscf.org > > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > > > > o ------------------------------------------------ o > > | Gabriele Sclauzero, PhD > > Student | > > | c/o: SISSA & CNR-INFM > > Democritos, | > > | via Beirut 2-4, > > 34014 Trieste (Italy) | > > | email: > > sclauzer at sissa.it | > > | phone: +39 040 3787 > > 511 | > > | skype: > > gurlonotturno | > > o ------------------------------------------------ o > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > Pieremanuele Canepa > Room 230 > School of Physical Sciences, Ingram Building, > University of Kent, Canterbury, Kent, > CT2 7NH > United Kingdom > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From akohlmey at cmm.chem.upenn.edu Mon May 25 17:54:13 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 May 2009 11:54:13 -0400 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> <443262146.09322@mail.sdu.edu.cn> Message-ID: <1243266853.10549.9052.camel@zero> On Mon, 2009-05-25 at 23:05 +0800, Hou Dong wrote: it is _bad_ advice to mix dynamical and shared libraries that contain the same code. please everybody disregard this post. thanks, axel. > > > From: pc229 at kent.ac.uk > Sent: Monday, May 25, 2009 10:51 PM > To: PWSCF Forum > Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to > optimize the U > > >Anyway I tried both the static and the dynamics compilation using > either this string > >ifort -o resp_mat.x resp_mat.f90 \ > >-L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lguide > -lpthread > > try -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 > -lmkl_core -lguide -lpthread > >and the output is : > .... > >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > reference to `mkl_serv_load_fun' > >/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined > reference to `mkl_serv_load_dll' > > it is related to a new feature of Intel MKL version 10 and higher. > add -lmkl_core may solve these problems. > > > ------------------------------------ > Dong Hou > School of Physics , Shandong University > Jinan, Shandong ,P.R. China. > houdong_phy at mail.sdu.edu.cn > --------------------------------------------- > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From sclauzer at sissa.it Mon May 25 18:00:25 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Mon, 25 May 2009 18:00:25 +0200 Subject: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U In-Reply-To: <1243266715.10549.9045.camel@zero> References: <8b443ba10905250414ifb451a0u246933620a15d6bc@mail.gmail.com> <4A1AA89A.8010701@sissa.it> <1243266715.10549.9045.camel@zero> Message-ID: <4A1AC099.5000007@sissa.it> Axel Kohlmeyer wrote: > On Mon, 2009-05-25 at 15:48 +0100, pc229 at kent.ac.uk wrote: >> I do apologize, >> I made a mistake when I paste the string that I use to compile it. >> Anyway I tried both the static and the dynamics compilation using >> either this string >> >> and the output is : >> >> >> and >> ifort -o resp_mat.x resp_mat.f90 \ >> -L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack64 -lmkl -lguide >> -lpthread > > why suddenly use now -lmkl_lapack64 ??? > there is not libmkl_lapack64.so in the 32bit version of mkl. In my machine at SISSA there is, but maybe it is a rather exotic installation... (actually I was surprised myself...) [sclauzer at brenta:/usr/local/intel_mkl_90/lib/32]$ ls -l total 110268 -r--r--r-- 1 root root 267450 Oct 6 2006 libguide.a -r-xr-xr-x 1 root root 226328 Oct 6 2006 libguide.so -r-xr-xr-x 1 root root 4744646 Oct 6 2006 libmkl_def.so -r--r--r-- 1 root root 47116850 Oct 6 2006 libmkl_ia32.a -r-xr-xr-x 1 root root 663086 Oct 6 2006 libmkl_ias.so -r-xr-xr-x 1 root root 2967647 Oct 6 2006 libmkl_lapack32.so -r-xr-xr-x 1 root root 3014308 Oct 6 2006 libmkl_lapack64.so -r--r--r-- 1 root root 10418470 Oct 6 2006 libmkl_lapack.a -r-xr-xr-x 1 root root 5645669 Oct 6 2006 libmkl_p3.so -r-xr-xr-x 1 root root 6511253 Oct 6 2006 libmkl_p4m.so -r-xr-xr-x 1 root root 6462455 Oct 6 2006 libmkl_p4p.so -r-xr-xr-x 1 root root 6586091 Oct 6 2006 libmkl_p4.so -r-xr-xr-x 1 root root 768161 Oct 6 2006 libmkl.so -r--r--r-- 1 root root 1333138 Oct 6 2006 libmkl_solver.a -r-xr-xr-x 1 root root 2189200 Oct 6 2006 libmkl_vml_def.so -r-xr-xr-x 1 root root 2595590 Oct 6 2006 libmkl_vml_p3.so -r-xr-xr-x 1 root root 3623813 Oct 6 2006 libmkl_vml_p4m.so -r-xr-xr-x 1 root root 3624853 Oct 6 2006 libmkl_vml_p4p.so -r-xr-xr-x 1 root root 3640224 Oct 6 2006 libmkl_vml_p4.so -r-xr-xr-x 1 root root 476207 Oct 6 2006 libvml.so In fact libmkl_lapack32.so contains single precision LAPACK routines, while libmkl_lapack64.so the double precision ones. Anyway I agree that the better way to understand how to link MKL is to read the local guide or check Intel's website. GS > it should work with plain -lmkl_lapack > > cheers, > axel. > > >> I cannot manage this problem. If it doesn't matter, can you please >> send me, privately, your compiled version? I am struggling a lot. >> Thank you, Piero > > _please_ rather then doing random stuff that does not make sense, > spend some time reading the MKL documentation. it is quite detailed, > useful and explains _everything_ about linking MKL dynamically and > statically. you have to keep in mind that the linux linker always > prefers shared libraries (.so) over static libraries. you should > also see from the documentation that the static version or lapack > is already included in libmkl_ia32.a. in conclusion: either link with > > -openmp -lmkl_lapack -lmkl > > or > > -i-static -openmp -lmkl_ia32 > > or > > -lmkl_lapack -lmkl_intel -lmkl_sequential -lmkl_core > > cheers, > axel. > >> ----- Original Message ----- >> From: Gabriele Sclauzero >> Date: Monday, May 25, 2009 3:18 pm >> Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to >> optimize the U >> To: PWSCF Forum >> >>> I thought Axel already replied to this post... >>> >>> Pieremanuele Canepa wrote: >>>> Dear all, >>>> I'm trying to compile the comp_resp_mat.j concerning the >>> DFT+U. Then, I >>>> tried to use the default settings (ifort-o r.x resp_mat.f90 >>>> -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 - >>> lguide >>>> -lpthread) and what I got, is reported below. Do you have any >>> suggestions?> Probably I am making some stupid mistake. >>> have somebody already >>>> compiled this program on 32 bit machine? >>>> I don't have any experience compiling with lapack libraries as >>> you can >>>> probably realize. >>>> Cheers, Piero >>> I don't understand if you're trying to link against mkl 7.2 or >>> 10.1. Do you know at least >>> which version is installed on your machine and where? Try to >>> find out this first. >>> >>> On my 32 bit Intel machine I've got MKL 9.0 installed in >>> /usr/local/intel_mkl_90/ and this >>> compilation command works (for static linking): >>> >>> ifort -o resp_mat.x resp_mat.f90 \ >>> -L"/usr/local/intel_mkl_90/lib/32" - >>> lmkl_lapack -lmkl_ia32 -lguide -lpthread >>> >>> ldd resp_mat.x >>> linux-gate.so.1 >>> => (0x00af3000) >>> libguide.so => >>> /usr/local/intel_mkl_90/lib/32/libguide.so (0x00a55000) >>> libm.so.6 => >>> /lib/libm.so.6 (0x00aa4000) >>> libc.so.6 => >>> /lib/libc.so.6 (0x00111000) >>> libgcc_s.so.1 >>> => /lib/libgcc_s.so.1 (0x00d4c000) >>> libdl.so.2 => >>> /lib/libdl.so.2 (0x00acd000) >>> libpthread.so.0 >>> => /lib/libpthread.so.0 (0x00ad3000) >>> /lib/ld- >>> linux.so.2 (0x00941000) >>> >>> >>> If you want to link dynamically use: >>> >>> ifort -o resp_mat.x resp_mat.f90 \ >>> -L"/usr/local/intel_mkl_90/lib/32" - >>> lmkl_lapack64 -lmkl -lguide -lpthread >>> >>> ldd resp_mat.x >>> linux-gate.so.1 >>> => (0x00741000) >>> >>> libmkl_lapack64.so => >>> /usr/local/intel_mkl_90/lib/32/libmkl_lapack64.so (0x00c32000) >>> libmkl.so => >>> /usr/local/intel_mkl_90/lib/32/libmkl.so (0x00eba000) >>> libguide.so => >>> /usr/local/intel_mkl_90/lib/32/libguide.so (0x00111000) >>> libpthread.so.0 >>> => /lib/libpthread.so.0 (0x00ad3000) >>> libm.so.6 => >>> /lib/libm.so.6 (0x00aa4000) >>> libc.so.6 => >>> /lib/libc.so.6 (0x0095f000) >>> libgcc_s.so.1 >>> => /lib/libgcc_s.so.1 (0x0015e000) >>> libdl.so.2 => >>> /lib/libdl.so.2 (0x00acd000) >>> /lib/ld- >>> linux.so.2 (0x00941000) >>> >>> Obviously you must change the path to the exact location of your >>> MKL installation >>> >>> >>> HTH >>> >>> GS >>> >>> >>>> resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. >>>> resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. >>>> resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED. >>>> resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED. >>>> resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED. >>>> resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED. >>>> /tmp/ifortMUANg9.o: In function `invmat_.': >>>> resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_' >>>> resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_' >>>> resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_' >>>> /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so >>>> : undefined >>> reference to >>>> `mkl_pds_c_blklu_unsym_omp_pardiso' >>>> /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so >>>> : undefined >>> reference to >>>> `mkl_lapack_cpptrs' >>>> /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so >>>> : undefined >>> reference to >>>> `mkl_serv_strcat2' >>>> /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so >>>> : undefined >>> reference to >>>> `mkl_pds_blklu_omp_pardiso' >>>> and so on >>>> -- >>>> Pieremanuele Canepa >>>> Room 230 >>>> School of Physical Sciences, Ingram Building, >>>> University of Kent, Canterbury, Kent, >>>> CT2 7NH >>>> United Kingdom >>>> ----------------------------------------------------------- >>>> >>>> >>>> --------------------------------------------------------------- >>> --------- >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> Pw_forum at pwscf.org >>>> http://www.democritos.it/mailman/listinfo/pw_forum >>> -- >>> >>> >>> o ------------------------------------------------ o >>> | Gabriele Sclauzero, PhD >>> Student | >>> | c/o: SISSA & CNR-INFM >>> Democritos, | >>> | via Beirut 2-4, >>> 34014 Trieste (Italy) | >>> | email: >>> sclauzer at sissa.it | >>> | phone: +39 040 3787 >>> 511 | >>> | skype: >>> gurlonotturno | >>> o ------------------------------------------------ o >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >>> >> --- >> Pieremanuele Canepa >> Room 230 >> School of Physical Sciences, Ingram Building, >> University of Kent, Canterbury, Kent, >> CT2 7NH >> United Kingdom >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sophia.nishad at gmail.com Mon May 25 18:33:28 2009 From: sophia.nishad at gmail.com (Sophia Nishad) Date: Mon, 25 May 2009 11:33:28 -0500 Subject: [Pw_forum] Optical phonon convergence at Gamma-point In-Reply-To: <1243266065.10549.9013.camel@zero> References: <8a0137e60905230817m506a597cg4e7f377c74e53343@mail.gmail.com> <8a0137e60905241404j38b11858w5f8a4825a3e204a7@mail.gmail.com> <5382CB05-00DC-462B-8F1A-91D299697528@democritos.it> <8a0137e60905250621v3933422obc885a644e5dfbb0@mail.gmail.com> <1243266065.10549.9013.camel@zero> Message-ID: <8a0137e60905250933s598d7ccy28853df425ae4e23@mail.gmail.com> Thank you, Prof. Baroni, and Alex. Sophia UCB, USA. On Mon, May 25, 2009 at 10:41 AM, Axel Kohlmeyer < akohlmey at cmm.chem.upenn.edu> wrote: > On Mon, 2009-05-25 at 06:21 -0700, Sophia Nishad wrote: > > > > Thanks, Paolo. I have little confusion about your comment though. > > > > Earlier I had used norm-conserving PP in supercell calculations as > > well; that produced converged frequency of 518.2 for Si. So, is it > > possible that linear response calculations might not give better > > results compared to supercell calculations (which is very expensive). > > The result could be sensitive to the assumptions of linear response > > theory as well?? > > no. there also is the error induced by using the electron density and > the potential on a finite grid. those are somewhat random, depending > on choice of k-points and atom positions and availability of FFT grids. > if you get less than 2% deviation from experiment then this is rather > by chance than by constructions. > > cheers, > axel. > > > > > Thanks, > > Sophia > > > > > > On Mon, May 25, 2009 at 1:26 AM, Paolo Giannozzi > > wrote: > > > > On May 24, 2009, at 23:04 , Sophia Nishad wrote: > > > > > it is still off from the experiemntal value by ~8cm-1. > > > > > > both pseudopotentials and approximations to the > > exchange-correlation > > functional introduce intrinsic errors. It is quite hard to get > > any > > better > > results than those you get. > > > > Paolo > > --- > > Paolo Giannozzi, Democritos and University of Udine, Italy > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > > > > -- > > Sophia > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > ======================================================================= > If you make something idiot-proof, the universe creates a better idiot. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Sophia -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090525/3cf8db8c/attachment-0001.htm From ismotochi at gmail.com Mon May 25 19:40:28 2009 From: ismotochi at gmail.com (isaac motochi) Date: Mon, 25 May 2009 05:40:28 -1200 Subject: [Pw_forum] Molybdenum pseudopotentials Message-ID: <9d6ea92e0905251040t464c378di993000dd3b472d0b@mail.gmail.com> Dear friends, Please can anyone sent me PBE pseudopotentials for molybdenum. Regards Isaac From akohlmey at cmm.chem.upenn.edu Mon May 25 22:16:20 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 25 May 2009 16:16:20 -0400 Subject: [Pw_forum] temperature In-Reply-To: <215224.61350.qm@web112505.mail.gq1.yahoo.com> References: <215224.61350.qm@web112505.mail.gq1.yahoo.com> Message-ID: <1243282580.10549.9645.camel@zero> On Mon, 2009-05-25 at 00:48 -0700, ali kazempoor wrote: dear ali, > Dear all > I want to study defect formation of charged defect at required > temperature . Since DFT is at T=0 , So I think that when we introduce what makes you say that DFT is at T=0K??? you are using the born-oppenheimer approximation and thus you can well study matter at elevated temperatures. however to get the properties of matter at temperatures != 0K you have to average over all statistically relevant configurations. please have a look into a book on statistical mechanics. > pressure on system it is equal to studying system at some finite > temperature. Is it true? and what is the relation and formula that please provide some references or at least some rationalizations to back up your claims. equations of state for solids usually relate the (potential) energy to the volume of a system (e.g. through the bulk modulus as a fairly constant material property). again, a look into a suitable textbook should help. cheers, axel. > shows how can we obtain the temperature by knowing the pressure of the > system? > thanks a lot > Ali Kazempour > Physics department, Isfahan University of Technology > 84156 Isfahan, Iran. Tel-1: +98 311 391 3733 > Fax: +98 311 391 2376 Tel-2: +98 311 391 2375 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From sanjeev0302 at rediffmail.com Tue May 26 06:32:08 2009 From: sanjeev0302 at rediffmail.com (Sanjeev K. Gupta) Date: 26 May 2009 04:32:08 -0000 Subject: [Pw_forum] pseudopotentials Message-ID: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> Dear PWSCF Users, Please can anyone sent me pseudopotentials for praseodymium, Promethium and Samarium. Regards Sanjeev Gupta S. K. Gupta PhD Candidate, Dept. of Physics, Bhavnagar University. Bhavnagar, 364 002., Gujarat. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/1812fd6e/attachment.htm From hqzhou at nju.edu.cn Tue May 26 06:58:03 2009 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Tue, 26 May 2009 12:58:03 +0800 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back References: <6E59C4FD0DDF4393A241193EC33E0EEC@ece.ncsu.edu> Message-ID: <94FAEF27BB9C431487E42DB71801D1BC@solarflare> Kostyantyn, The rule for transpose operation applies to not only the product of matrices, but also the product of a matrix and a vector. I believe Gabriele is right. Huiqun Zhou @Earth Sciences, Nanjing University, China 2Gabriele: I beleive there is a mistake in your first step. When you start with v_cart = A*v_cryst, you try to use the property (C*D)' = D' * C', which is true for the product of two matrices. In our case we have a product of matrix A and vector v_cryst. So you will get v_cart' = v_cryst' * A -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/dcf2f273/attachment.htm From baroni at sissa.it Tue May 26 07:22:05 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 26 May 2009 07:22:05 +0200 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back In-Reply-To: <94FAEF27BB9C431487E42DB71801D1BC@solarflare> References: <6E59C4FD0DDF4393A241193EC33E0EEC@ece.ncsu.edu> <94FAEF27BB9C431487E42DB71801D1BC@solarflare> Message-ID: <3E7B3D6C-B93A-4BC8-9A27-359FEA409F78@sissa.it> all in all, a "N vector" is nothing but a [Nx1] matrix. Its transpose (YES! it is well defined) is a [1xN] matrix. SB On May 26, 2009, at 6:58 AM, Huiqun Zhou wrote: > Kostyantyn? > > The rule for transpose operation applies to not only the product of > matrices, but also the > product of a matrix and a vector. > > I believe Gabriele is right. > > > Huiqun Zhou > @Earth Sciences, Nanjing University, China > > 2Gabriele: I beleive there is a mistake in your first step. When you > start with v_cart = A*v_cryst, you try to use the property (C*D)' = > D' * C', which is true for the product of two matrices. In our case > we have a product of matrix A and vector v_cryst. So you will get > v_cart' = v_cryst' * A > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/5e0fa9f3/attachment.htm From sclauzer at sissa.it Tue May 26 09:47:21 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 26 May 2009 09:47:21 +0200 Subject: [Pw_forum] pseudopotentials In-Reply-To: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> References: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> Message-ID: <4A1B9E89.9010305@sissa.it> Dear Sanjeev, even if someone has all three ready for you (though i doubt it could be possible), I hope you are well aware of the difficulties in threating such kind of elements (i.e. Lanthanides, which have open-shell 4f electrons in valence) within DFT. GS Sanjeev K. Gupta wrote: > Dear PWSCF Users, > > Please can anyone sent me pseudopotentials for praseodymium, Promethium > and Samarium. > > > Regards > Sanjeev Gupta > > S. K. Gupta > > PhD Candidate, > > Dept. of Physics, > > Bhavnagar University. > > Bhavnagar, 364 002., Gujarat. > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From cao at qtp.ufl.edu Tue May 26 10:05:11 2009 From: cao at qtp.ufl.edu (Chao Cao) Date: Tue, 26 May 2009 16:05:11 +0800 Subject: [Pw_forum] pseudopotentials In-Reply-To: <4A1B9E89.9010305@sissa.it> References: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> <4A1B9E89.9010305@sissa.it> Message-ID: <5AC5679C-E3F2-4117-804F-C3D8C0EE8680@qtp.ufl.edu> And we completely agree with Gabriele that 1. It is dangerous to deal with f-elements using DFT unless you know what you are doing 2. It is even more dangerous to deal with these elements with frozen-core approximations 3. These potentials are extremely difficult to generate. Best, Chao Cao Ph. D. in Physics Department of Physics and Quantum Theory Project University of Florida Gainesville, FL 32611 U. S. A. ? May 26, 2009?3:47 PM? Gabriele Sclauzero ??? > Dear Sanjeev, > > even if someone has all three ready for you (though i doubt it > could be possible), I > hope you are well aware of the difficulties in threating such kind > of elements (i.e. > Lanthanides, which have open-shell 4f electrons in valence) within > DFT. > > GS > > Sanjeev K. Gupta wrote: >> Dear PWSCF Users, >> >> Please can anyone sent me pseudopotentials for praseodymium, >> Promethium >> and Samarium. >> >> >> Regards >> Sanjeev Gupta >> >> S. K. Gupta >> >> PhD Candidate, >> >> Dept. of Physics, >> >> Bhavnagar University. >> >> Bhavnagar, 364 002., Gujarat. >> >> >> > > >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From cao at qtp.ufl.edu Tue May 26 09:59:30 2009 From: cao at qtp.ufl.edu (Chao Cao) Date: Tue, 26 May 2009 15:59:30 +0800 Subject: [Pw_forum] pseudopotentials In-Reply-To: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> References: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> Message-ID: Our group actually have a set of Lanthanide PAW data, although only La has been tested. We'll put them on-line once our paper is out. But I should warn you that none of the elements you requested is well- tested, so you could either wait for us to test them or you should do the tests. If you (or anyone else on this forum) need, please contact us, and we can talk about the details. Best, Chao Cao Ph. D. in Physics Department of Physics and Quantum Theory Project University of Florida Gainesville, FL 32611 U. S. A. ? May 26, 2009?12:32 PM? Sanjeev K. Gupta ??? > Dear PWSCF Users, > > Please can anyone sent me pseudopotentials for praseodymium, > Promethium and Samarium. > > > Regards > Sanjeev Gupta > > S. K. Gupta > > PhD Candidate, > > Dept. of Physics, > > Bhavnagar University. > > Bhavnagar, 364 002., Gujarat. > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/cd9000c3/attachment.htm From baroni at sissa.it Tue May 26 10:21:12 2009 From: baroni at sissa.it (Stefano Baroni) Date: Tue, 26 May 2009 10:21:12 +0200 Subject: [Pw_forum] pseudopotentials In-Reply-To: References: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> Message-ID: Hi! Please, consider using qe-forge for publishing and sharing this kind of data Cheers - SB On May 26, 2009, at 9:59 AM, Chao Cao wrote: > > Our group actually have a set of Lanthanide PAW data, although only > La has been tested. We'll put them on-line once our paper is out. > But I should warn you that none of the elements you requested is > well-tested, so you could either wait for us to test them or you > should do the tests. If you (or anyone else on this forum) need, > please contact us, and we can talk about the details. > > > > Best, > > Chao Cao > > Ph. D. in Physics > Department of Physics and Quantum Theory Project > University of Florida > Gainesville, FL 32611 > U. S. A. > > > > > ? May 26, 2009?12:32 PM? Sanjeev K. Gupta ??? > >> Dear PWSCF Users, >> >> Please can anyone sent me pseudopotentials for praseodymium, >> Promethium and Samarium. >> >> >> Regards >> Sanjeev Gupta >> >> S. K. Gupta >> >> PhD Candidate, >> >> Dept. of Physics, >> >> Bhavnagar University. >> >> Bhavnagar, 364 002., Gujarat. >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/d4efd14b/attachment.htm From cao at qtp.ufl.edu Tue May 26 13:14:06 2009 From: cao at qtp.ufl.edu (Chao Cao) Date: Tue, 26 May 2009 19:14:06 +0800 Subject: [Pw_forum] pseudopotentials In-Reply-To: References: <20090526043208.25148.qmail@f4mail-235-240.rediffmail.com> Message-ID: <986691D6-2255-4E25-B5A6-A8FEE32C09C5@qtp.ufl.edu> We will. But we have to do some extensive tests before doing that. :) Thanks for the nice work on PAW by the way. Best, Chao Cao Ph. D. Quantum Theory Project and Department of Physics University of Florida Gainesville, FL 32611 U. S. A. ? May 26, 2009?4:21 PM? Stefano Baroni ??? > Hi! Please, consider using qe-forge for publishing and sharing this > kind of data > Cheers - SB > > On May 26, 2009, at 9:59 AM, Chao Cao wrote: > >> >> Our group actually have a set of Lanthanide PAW data, although only >> La has been tested. We'll put them on-line once our paper is out. >> But I should warn you that none of the elements you requested is >> well-tested, so you could either wait for us to test them or you >> should do the tests. If you (or anyone else on this forum) need, >> please contact us, and we can talk about the details. >> >> >> >> Best, >> >> Chao Cao >> >> Ph. D. in Physics >> Department of Physics and Quantum Theory Project >> University of Florida >> Gainesville, FL 32611 >> U. S. A. >> >> >> >> >> ? May 26, 2009?12:32 PM? Sanjeev K. Gupta ??? >> >>> Dear PWSCF Users, >>> >>> Please can anyone sent me pseudopotentials for praseodymium, >>> Promethium and Samarium. >>> >>> >>> Regards >>> Sanjeev Gupta >>> >>> S. K. Gupta >>> >>> PhD Candidate, >>> >>> Dept. of Physics, >>> >>> Bhavnagar University. >>> >>> Bhavnagar, 364 002., Gujarat. >>> >>> >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://www.democritos.it/mailman/listinfo/pw_forum >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - > Trieste > http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / > stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une > morale de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/809f2b60/attachment-0001.htm From giannozz at democritos.it Tue May 26 17:00:37 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Tue, 26 May 2009 17:00:37 +0200 Subject: [Pw_forum] CONSTRAINT card In-Reply-To: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> References: <45070.152.14.74.117.1242939075.squirrel@webmail.ncsu.edu> Message-ID: <7DBD76A1-4C91-441C-8C40-4B02D853F254@democritos.it> On May 21, 2009, at 22:51 , Vivek Ranjan wrote: > I am wondering how to use CONSTRAINT card ? did you have a look at file Doc/constraints_HOWTO.tex? Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From dimpy.sharma at tyndall.ie Tue May 26 17:39:47 2009 From: dimpy.sharma at tyndall.ie (Dimpy Sharma) Date: Tue, 26 May 2009 16:39:47 +0100 Subject: [Pw_forum] scf calculation Message-ID: Hi Quantum espresso user, I have tried to perform a scf calculation using periodic boundary condition, however my calculation stops after first iteration step, is it because of overlapping of cell? can anybody suggest me ? The input file is given below: &CONTROL calculation ='scf' restart_mode ='from_scratch' outdir = '/sfiwork/dsharma/silane/pasq' pseudo_dir = '/sfiwork/dsharma/QE_Pseudos/' prefix = '1' tstress = .false. tprnfor = .true. etot_conv_thr = 1.D-4 forc_conv_thr = 1.D-3 nstep = 600 / &SYSTEM ibrav = 0 celldm(1) =7.680 nat = 94 ntyp = 3 ecutwfc = 40 ecutrho = 120 nbnd = 450 / &ELECTRONS diagonalization ='david' mixing_mode = 'plain' conv_thr = 1.0d-6 mixing_beta = 0.7 / CELL_PARAMETERS cubic 1.000000000 0.000000000 0.000000000 0.000000000 1.000000000 0.000000000 0.000000000 0.000000000 12.000000000 ATOMIC_SPECIES Si 28.08600 Si.pz-vbc.UPF H 1.00800 H.pz-vbc.UPF O 15.9994 O.pz-rrkjus.UPF ATOMIC_POSITIONS angstroms O 7.0781 6.9497 27.1761 O 4.3952 6.9992 27.2391 O 0.5652 4.5006 27.2372 O 3.2472 4.5482 27.1701 O 1.9438 0.0919 27.1630 O 5.7766 3.7444 27.1486 O 7.6277 1.8605 27.2622 O 3.7925 1.9787 27.2580 O 0.4478 0.5350 25.0261 O 3.4856 0.6080 25.0468 O 7.3297 3.2255 25.0426 O 4.2775 3.2996 25.0202 O 1.9513 3.9801 24.0193 O 5.7797 7.5116 24.0552 O 0.0040 5.7160 24.5904 O 3.8230 5.7805 24.5911 O 0.2056 7.2121 22.5056 O 3.6836 7.2984 22.5163 O 7.4958 4.2109 22.5160 O 4.0618 4.2895 22.5067 O 1.9285 0.3168 20.6977 O 5.7595 3.4671 20.6962 O 7.6593 1.8808 21.4706 O 3.7728 1.9700 21.4647 O 7.2855 7.3586 19.8746 O 4.2577 7.4328 19.8968 O 0.4187 4.0760 19.8952 O 3.4668 4.1529 19.8704 O 0.0019 1.9039 18.5887 O 3.8316 1.9347 18.5904 H 7.4599 6.0212 27.1358 H 5.3921 7.0051 27.1349 H 1.5615 4.4933 27.1304 H 3.6286 5.4768 27.1339 H 1.9154 6.4168 5.4186 H 5.7463 6.4168 5.4186 H 1.9154 2.5859 5.4186 H 5.7463 2.5859 5.4186 H 1.9154 1.2450 5.4186 H 5.7463 1.2450 5.4186 H 1.9154 5.0759 5.4186 H 5.7463 5.0759 5.4186 Si 3.4277 0.5123 26.6794 Si 7.2634 3.3235 26.6750 Si 0.4470 0.4413 26.6627 Si 4.2778 3.3947 26.6563 Si 3.5195 4.3107 24.0162 Si 7.3479 7.1869 24.0233 Si 0.3727 4.2589 24.0241 Si 4.2002 7.2396 24.0320 Si 3.4527 0.4290 21.1515 Si 7.2869 3.4090 21.1521 Si 0.3996 0.3623 21.1435 Si 4.2285 3.4744 21.1412 Si 7.5663 0.2867 18.3507 Si 3.9489 0.3177 18.3620 Si 0.1175 3.5208 18.3617 Si 3.7409 3.5519 18.3482 Si 1.9286 7.5732 17.2122 Si 5.7519 7.6559 16.9682 Si 1.9219 3.8428 16.9706 Si 5.7634 3.9249 17.2068 Si 1.9220 1.8288 15.7617 Si 5.7548 2.0065 15.7591 Si 1.9193 5.7818 15.7231 Si 5.7524 5.7178 15.7201 Si 0.0066 1.9168 14.4083 Si 3.8367 1.9161 14.4106 Si 0.0050 5.7504 14.3813 Si 3.8343 5.7461 14.3845 Si 7.6487 0.0094 13.0498 Si 3.8488 0.0070 13.0529 Si 0.0177 3.8258 13.0511 Si 3.8171 3.8244 13.0534 Si 1.9180 0.0028 11.7170 Si 5.7477 7.6614 11.6950 Si 1.9180 3.8318 11.6954 Si 5.7469 3.8289 11.7177 Si 1.9154 1.9154 10.3635 Si 5.7463 1.9154 10.3635 Si 1.9154 5.7463 10.3635 Si 5.7463 5.7463 10.3635 Si 0.0000 1.9154 9.0087 Si 3.8309 1.9154 9.0087 Si 0.0000 5.7463 9.0087 Si 3.8309 5.7463 9.0087 Si 0.0000 0.0000 7.6539 Si 3.8309 0.0000 7.6539 Si 0.0000 3.8309 7.6539 Si 3.8309 3.8309 7.6539 Si 1.9154 0.0000 6.2994 Si 5.7463 0.0000 6.2994 Si 1.9154 3.8309 6.2994 Si 5.7463 3.8309 6.2994 K_POINTS automatic 4 1 1 0 0 0 Thanks Dimpy -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/d26c5bcd/attachment.htm From sclauzer at sissa.it Tue May 26 17:44:48 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Tue, 26 May 2009 17:44:48 +0200 Subject: [Pw_forum] scf calculation In-Reply-To: References: Message-ID: <4A1C0E70.4020502@sissa.it> Dimpy Sharma wrote: > > Hi Quantum espresso user, > > I have tried to perform a scf calculation using periodic boundary > condition, however my calculation stops after first iteration step, is > it because of overlapping of cell? can anybody suggest me ? The input I've never heard about "overlapping of cell". Anyway, I suggest to correct your atomic positions in order not to have overlapping atoms and then retry (if you're not sure of your coordinates consider using a visualization tool such as XCrysDen). GS > file is given below: > > > &CONTROL > calculation ='scf' > restart_mode ='from_scratch' > outdir = '/sfiwork/dsharma/silane/pasq' > pseudo_dir = '/sfiwork/dsharma/QE_Pseudos/' > prefix = '1' > tstress = .false. > tprnfor = .true. > etot_conv_thr = 1.D-4 > forc_conv_thr = 1.D-3 > nstep = 600 > / > &SYSTEM > ibrav = 0 > celldm(1) =7.680 > nat = 94 > ntyp = 3 > ecutwfc = 40 > ecutrho = 120 > nbnd = 450 > / > &ELECTRONS > diagonalization ='david' > mixing_mode = 'plain' > conv_thr = 1.0d-6 > mixing_beta = 0.7 > / > CELL_PARAMETERS cubic > > 1.000000000 0.000000000 0.000000000 > 0.000000000 1.000000000 0.000000000 > 0.000000000 0.000000000 12.000000000 > ATOMIC_SPECIES > Si 28.08600 Si.pz-vbc.UPF > H 1.00800 H.pz-vbc.UPF > O 15.9994 O.pz-rrkjus.UPF > ATOMIC_POSITIONS angstroms > O 7.0781 6.9497 27.1761 > O 4.3952 6.9992 27.2391 > O 0.5652 4.5006 27.2372 > O 3.2472 4.5482 27.1701 > O 1.9438 0.0919 27.1630 > O 5.7766 3.7444 27.1486 > O 7.6277 1.8605 27.2622 > O 3.7925 1.9787 27.2580 > O 0.4478 0.5350 25.0261 > O 3.4856 0.6080 25.0468 > O 7.3297 3.2255 25.0426 > O 4.2775 3.2996 25.0202 > O 1.9513 3.9801 24.0193 > O 5.7797 7.5116 24.0552 > O 0.0040 5.7160 24.5904 > O 3.8230 5.7805 24.5911 > O 0.2056 7.2121 22.5056 > O 3.6836 7.2984 22.5163 > O 7.4958 4.2109 22.5160 > O 4.0618 4.2895 22.5067 > O 1.9285 0.3168 20.6977 > O 5.7595 3.4671 20.6962 > O 7.6593 1.8808 21.4706 > O 3.7728 1.9700 21.4647 > O 7.2855 7.3586 19.8746 > O 4.2577 7.4328 19.8968 > O 0.4187 4.0760 19.8952 > O 3.4668 4.1529 19.8704 > O 0.0019 1.9039 18.5887 > O 3.8316 1.9347 18.5904 > H 7.4599 6.0212 27.1358 > H 5.3921 7.0051 27.1349 > H 1.5615 4.4933 27.1304 > H 3.6286 5.4768 27.1339 > H 1.9154 6.4168 5.4186 > H 5.7463 6.4168 5.4186 > H 1.9154 2.5859 5.4186 > H 5.7463 2.5859 5.4186 > H 1.9154 1.2450 5.4186 > H 5.7463 1.2450 5.4186 > H 1.9154 5.0759 5.4186 > H 5.7463 5.0759 5.4186 > Si 3.4277 0.5123 26.6794 > Si 7.2634 3.3235 26.6750 > Si 0.4470 0.4413 26.6627 > Si 4.2778 3.3947 26.6563 > Si 3.5195 4.3107 24.0162 > Si 7.3479 7.1869 24.0233 > Si 0.3727 4.2589 24.0241 > Si 4.2002 7.2396 24.0320 > Si 3.4527 0.4290 21.1515 > Si 7.2869 3.4090 21.1521 > Si 0.3996 0.3623 21.1435 > Si 4.2285 3.4744 21.1412 > Si 7.5663 0.2867 18.3507 > Si 3.9489 0.3177 18.3620 > Si 0.1175 3.5208 18.3617 > Si 3.7409 3.5519 18.3482 > Si 1.9286 7.5732 17.2122 > Si 5.7519 7.6559 16.9682 > Si 1.9219 3.8428 16.9706 > Si 5.7634 3.9249 17.2068 > Si 1.9220 1.8288 15.7617 > Si 5.7548 2.0065 15.7591 > Si 1.9193 5.7818 15.7231 > Si 5.7524 5.7178 15.7201 > Si 0.0066 1.9168 14.4083 > Si 3.8367 1.9161 14.4106 > Si 0.0050 5.7504 14.3813 > Si 3.8343 5.7461 14.3845 > Si 7.6487 0.0094 13.0498 > Si 3.8488 0.0070 13.0529 > Si 0.0177 3.8258 13.0511 > Si 3.8171 3.8244 13.0534 > Si 1.9180 0.0028 11.7170 > Si 5.7477 7.6614 11.6950 > Si 1.9180 3.8318 11.6954 > Si 5.7469 3.8289 11.7177 > Si 1.9154 1.9154 10.3635 > Si 5.7463 1.9154 10.3635 > Si 1.9154 5.7463 10.3635 > Si 5.7463 5.7463 10.3635 > Si 0.0000 1.9154 9.0087 > Si 3.8309 1.9154 9.0087 > Si 0.0000 5.7463 9.0087 > Si 3.8309 5.7463 9.0087 > Si 0.0000 0.0000 7.6539 > Si 3.8309 0.0000 7.6539 > Si 0.0000 3.8309 7.6539 > Si 3.8309 3.8309 7.6539 > Si 1.9154 0.0000 6.2994 > Si 5.7463 0.0000 6.2994 > Si 1.9154 3.8309 6.2994 > Si 5.7463 3.8309 6.2994 > K_POINTS automatic > 4 1 1 0 0 0 > > Thanks > > Dimpy > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From Giovanni.Cantele at na.infn.it Tue May 26 18:02:48 2009 From: Giovanni.Cantele at na.infn.it (Giovanni Cantele) Date: Tue, 26 May 2009 18:02:48 +0200 Subject: [Pw_forum] scf calculation In-Reply-To: References: Message-ID: <4A1C12A8.2070102@na.infn.it> Dimpy Sharma wrote: > > Hi Quantum espresso user, > > I have tried to perform a scf calculation using periodic boundary > condition, however my calculation stops after first iteration step, is > it because of overlapping of cell? can anybody suggest me ? The input > file is given below: > > > &CONTROL > calculation ='scf' > restart_mode ='from_scratch' > outdir = '/sfiwork/dsharma/silane/pasq' > pseudo_dir = '/sfiwork/dsharma/QE_Pseudos/' > prefix = '1' > tstress = .false. > tprnfor = .true. > etot_conv_thr = 1.D-4 > forc_conv_thr = 1.D-3 > nstep = 600 > / > &SYSTEM > ibrav = 0 > celldm(1) =7.680 > nat = 94 > ntyp = 3 > ecutwfc = 40 > ecutrho = 120 > nbnd = 450 > / > &ELECTRONS > diagonalization ='david' > mixing_mode = 'plain' > conv_thr = 1.0d-6 > mixing_beta = 0.7 > / > CELL_PARAMETERS cubic > > 1.000000000 0.000000000 0.000000000 > 0.000000000 1.000000000 0.000000000 > 0.000000000 0.000000000 12.000000000 > ATOMIC_SPECIES > Si 28.08600 Si.pz-vbc.UPF > H 1.00800 H.pz-vbc.UPF > O 15.9994 O.pz-rrkjus.UPF > ATOMIC_POSITIONS angstroms > O 7.0781 6.9497 27.1761 > O 4.3952 6.9992 27.2391 > O 0.5652 4.5006 27.2372 > O 3.2472 4.5482 27.1701 > O 1.9438 0.0919 27.1630 > O 5.7766 3.7444 27.1486 > O 7.6277 1.8605 27.2622 > O 3.7925 1.9787 27.2580 > O 0.4478 0.5350 25.0261 > O 3.4856 0.6080 25.0468 > O 7.3297 3.2255 25.0426 > O 4.2775 3.2996 25.0202 > O 1.9513 3.9801 24.0193 > O 5.7797 7.5116 24.0552 > O 0.0040 5.7160 24.5904 > O 3.8230 5.7805 24.5911 > O 0.2056 7.2121 22.5056 > O 3.6836 7.2984 22.5163 > O 7.4958 4.2109 22.5160 > O 4.0618 4.2895 22.5067 > O 1.9285 0.3168 20.6977 > O 5.7595 3.4671 20.6962 > O 7.6593 1.8808 21.4706 > O 3.7728 1.9700 21.4647 > O 7.2855 7.3586 19.8746 > O 4.2577 7.4328 19.8968 > O 0.4187 4.0760 19.8952 > O 3.4668 4.1529 19.8704 > O 0.0019 1.9039 18.5887 > O 3.8316 1.9347 18.5904 > H 7.4599 6.0212 27.1358 > H 5.3921 7.0051 27.1349 > H 1.5615 4.4933 27.1304 > H 3.6286 5.4768 27.1339 > H 1.9154 6.4168 5.4186 > H 5.7463 6.4168 5.4186 > H 1.9154 2.5859 5.4186 > H 5.7463 2.5859 5.4186 > H 1.9154 1.2450 5.4186 > H 5.7463 1.2450 5.4186 > H 1.9154 5.0759 5.4186 > H 5.7463 5.0759 5.4186 > Si 3.4277 0.5123 26.6794 > Si 7.2634 3.3235 26.6750 > Si 0.4470 0.4413 26.6627 > Si 4.2778 3.3947 26.6563 > Si 3.5195 4.3107 24.0162 > Si 7.3479 7.1869 24.0233 > Si 0.3727 4.2589 24.0241 > Si 4.2002 7.2396 24.0320 > Si 3.4527 0.4290 21.1515 > Si 7.2869 3.4090 21.1521 > Si 0.3996 0.3623 21.1435 > Si 4.2285 3.4744 21.1412 > Si 7.5663 0.2867 18.3507 > Si 3.9489 0.3177 18.3620 > Si 0.1175 3.5208 18.3617 > Si 3.7409 3.5519 18.3482 > Si 1.9286 7.5732 17.2122 > Si 5.7519 7.6559 16.9682 > Si 1.9219 3.8428 16.9706 > Si 5.7634 3.9249 17.2068 > Si 1.9220 1.8288 15.7617 > Si 5.7548 2.0065 15.7591 > Si 1.9193 5.7818 15.7231 > Si 5.7524 5.7178 15.7201 > Si 0.0066 1.9168 14.4083 > Si 3.8367 1.9161 14.4106 > Si 0.0050 5.7504 14.3813 > Si 3.8343 5.7461 14.3845 > Si 7.6487 0.0094 13.0498 > Si 3.8488 0.0070 13.0529 > Si 0.0177 3.8258 13.0511 > Si 3.8171 3.8244 13.0534 > Si 1.9180 0.0028 11.7170 > Si 5.7477 7.6614 11.6950 > Si 1.9180 3.8318 11.6954 > Si 5.7469 3.8289 11.7177 > Si 1.9154 1.9154 10.3635 > Si 5.7463 1.9154 10.3635 > Si 1.9154 5.7463 10.3635 > Si 5.7463 5.7463 10.3635 > Si 0.0000 1.9154 9.0087 > Si 3.8309 1.9154 9.0087 > Si 0.0000 5.7463 9.0087 > Si 3.8309 5.7463 9.0087 > Si 0.0000 0.0000 7.6539 > Si 3.8309 0.0000 7.6539 > Si 0.0000 3.8309 7.6539 > Si 3.8309 3.8309 7.6539 > Si 1.9154 0.0000 6.2994 > Si 5.7463 0.0000 6.2994 > Si 1.9154 3.8309 6.2994 > Si 5.7463 3.8309 6.2994 > K_POINTS automatic > 4 1 1 0 0 0 > > Thanks > > Dimpy > If you try to use a visualization tool like XCrysDen, you'll discover that many atoms are (almost) overlapped. An example: atoms 7 - 8 O 7.6277 1.8605 27.2622 O 3.7925 1.9787 27.2580 You specify the atomic positions in Angstrom. Because you specify both CELL_PARAMETERS and celldm(1) (which is in atomic units), the unit cell size along x is 1.0 * 7.680 * 0.529177 A = 4.064 A. Therefore the x coordinate of the first atom is equivalent to 7.6277-4.064 A = 3.563 A. Therefore the distance between the two atoms is only .26 A. Similar considerations for many more atom pairs. More hints: - not very sure about the setting: ecutwfc = 40 ecutrho = 120 ecutrho should be at least 4*ecutwfc, larger (6-12 times ecutrho) when using US pseudo - you choose a 4x1x1 grid, being a=b maybe one expects that the same k-point sampling is needed along both x and y directions Giovanni -- Dr. Giovanni Cantele Coherentia CNR-INFM and Dipartimento di Scienze Fisiche Universita' di Napoli "Federico II" Complesso Universitario di Monte S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy Phone: +39 081 676910 Fax: +39 081 676346 E-mail: giovanni.cantele at cnr.it giovanni.cantele at na.infn.it Web: http://people.na.infn.it/~cantele Research Group: http://www.nanomat.unina.it Skype contact: giocan74 From cristian.degliesposti at unibo.it Tue May 26 18:15:45 2009 From: cristian.degliesposti at unibo.it (Cristian Degli Esposti Boschi) Date: Tue, 26 May 2009 18:15:45 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro Message-ID: <4A1C15B1.5010902@unibo.it> >That sounds strange... even if you build a mpif90 on top of gfortran >(for instance using >OpenMPI), you should call it as "mpif90", not "gfortran". >I don't understand if you want to compile a serial or a parallel >version. If you want a serial one, use >./configure --disable-parallel Ok, in order to simplify the matter I restrict myself to a serial build. Therefore I pass the option "--disable-parallel" to configure and make all (using gfortran). I tried both pw.x and cp.x with two input files for a rather demanding system; on a 2 Xeon Quad-core machine with 16 GB RAM very similar to the Mac Pro they are successfully processed while on the Mac Pro I get: pw.x(45534) malloc: *** mmap(size=1093816320) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Operating system error: Cannot allocate memory Out of memory or cp.x(45549) malloc: *** mmap(size=546910208) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Operating system error: Cannot allocate memory Out of memory One thing I noticed in the file config.log is that there is no malloc usability. The file malloc.h is not found, so I made a symlink from /usr/include/malloc/malloc.h to /usr/include/malloc.h. But then the same problem happens with "struct mallinfo.arena"... Is there a special variable to pass to locate correctly the (malloc related) include files under OSX Leopard 10.5, maybe when I use the compilers from FINK? Finally, what I meant about 64-bit compiling is that I used to get similar errors with a C++ code of mine: dmrgcc(45305) malloc: *** mmap(size=487424) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug terminate called after throwing an instance of 'std::bad_alloc' what(): St9bad_alloc In that case I solved the problem by passing explicitely the option "-m64" to the compilers; with Quantum Espresso the configure step with -m64 is ok, but when I make all I get several errors in fft_scalar. Thanks for your time. Cristian -- ___________________________________________________ Cristian Degli Esposti Boschi CNR, CNISM, Unita' di Ricerca di Bologna, c/o Dipartimento di Fisica, Universita' di Bologna viale Berti-Pichat, 6/2, 40127, Bologna, Italia tel. ++39 051 2095114 fax ++39 051 2095113 e-mail: cristian.degliesposti -AT- unibo.it web: http://www.df.unibo.it/fismat/theory ___________________________________________________ From wshen02 at gmail.com Tue May 26 21:34:43 2009 From: wshen02 at gmail.com (=?ISO-2022-JP?B?GyRCSjhEQBsoQg==?=) Date: Tue, 26 May 2009 12:34:43 -0700 Subject: [Pw_forum] Hello, everybody, questions about the calculation about the rare earth material Dy Message-ID: <2c9df8a80905261234v56d925f7j47be60d8ea31e40d@mail.gmail.com> Dear all, I am currently doing the calculation about the rare earth material Dy, and since I can't find the prevailing pseudopotential for Dy, I generate Dy pseudopotential by myself,with a ld1.x input file as follows &input title='Dy', zed=66.0, rel=2, iswitch=3, rlderiv=2.50, eminld=-5.0, emaxld=5.0, deld=0.02, nld=5, config='[Xe] 4F10 6S2 5D0 6P0 ' iswitch=3, dft='LDA', / &inputp pseudotype=3, lloc=0, file_pseudopw='Dy.LDA.UPF', nlcc=.true., rcore=3.91516, rcloc=5 / 9 4F 4 3 6.00 0.00 2.10 2.40 2.50 4F 4 3 0.00 -0.20 2.10 2.40 2.50 4F 4 3 4.00 0.00 2.10 2.40 3.50 4F 4 3 0.00 -0.20 2.10 2.40 3.50 5D 3 2 0.00 0.00 1.90 2.30 2.50 5D 3 2 0.00 -0.20 1.90 2.30 2.50 5D 3 2 0.00 0.00 1.90 2.30 1.50 5D 3 2 0.00 -0.20 1.90 2.30 1.50 6S 1 0 2.00 0.00 2.60 2.60 0.50 With this input file. I can definitely get the pseudopotential but with some warnings about Zero norm: self consistency problem; state: 2 (l= 3, j=3.5) Zero norm: self consistency problem; state: 1 (l= 3, j=2.5) but when I tried to calculate the band structure for the hexagonal Dy lattice, I can't get the right band structure. I searched on the pwscf website and I find it says, If X is one of the rare earths: please consider first if DFT is suitable for your system! SO can I ask what is the right system for the rare earth calculation? and if somepeople have the experience about the rare earth pseudopotential calculation, please give me some advices about the pseudopotential generating!! Thanks a lot Wen Shen Ph. D candidate Physics department Georgetown University -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/e46a7c05/attachment.htm From kboryse at ncsu.edu Tue May 26 22:10:42 2009 From: kboryse at ncsu.edu (Kostyantyn Borysenko) Date: Tue, 26 May 2009 16:10:42 -0400 Subject: [Pw_forum] Coordinate transorms: Crystal to Cartesian and back In-Reply-To: <2c9df8a80905261234v56d925f7j47be60d8ea31e40d@mail.gmail.com> References: <2c9df8a80905261234v56d925f7j47be60d8ea31e40d@mail.gmail.com> Message-ID: <1D9174197C2D48AC8BB16F8D1C8EA920@ece.ncsu.edu> Thanks everyone! Indeed, I was wrong about transposing the product of matrix and vector. Silly mistake on my part. Like I said in my previous message, I finally was able to reproduce the expressions in those two subroutines. What I didn't understand at first is that if a matrix element depends on wave vectors k and q (like in our case: ) , the transition between cartesian and crystal coordinates actually refers to transition between basis of Cartesian unit vectors and the basis of unit vectors in reciprocal lattice. So it is a crystal - but in reciprocal space! Thanks again for your help. Best, Kostyantyn Borysenko Department of Electrical and Computer Engineering NCSU > all in all, a "N vector" is nothing but a [Nx1] matrix. Its transpose > (YES! it is well defined) is a [1xN] matrix. SB > On May 26, 2009, at 6:58 AM, Huiqun Zhou wrote: > Kostyantyn , > > The rule for transpose operation applies to not only the product of > matrices, but also the > product of a matrix and a vector. > > I believe Gabriele is right. > > > Huiqun Zhou > @Earth Sciences, Nanjing University, China > > 2Gabriele: I beleive there is a mistake in your first step. When you > start with v_cart = A*v_cryst, you try to use the property (C*D)' = > D' * C', which is true for the product of two matrices. In our case > we have a product of matrix A and vector v_cryst. So you will get > v_cart' = v_cryst' * A > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090526/fad1f71c/attachment.htm From sclauzer at sissa.it Wed May 27 08:50:23 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Wed, 27 May 2009 08:50:23 +0200 Subject: [Pw_forum] Hello, everybody, questions about the calculation about the rare earth material Dy In-Reply-To: <2c9df8a80905261234v56d925f7j47be60d8ea31e40d@mail.gmail.com> References: <2c9df8a80905261234v56d925f7j47be60d8ea31e40d@mail.gmail.com> Message-ID: <4A1CE2AF.50306@sissa.it> ?? wrote: > Dear all, > I am currently doing the calculation about the rare earth material Dy, > and since I can't find the prevailing pseudopotential for Dy, I generate > and if somepeople have the experience about the rare earth > pseudopotential calculation, please give me some > advices about the pseudopotential generating!! Please have a look at the forum archives. There is a post of just few days ago (maybe yesterday) about rare earth PAW datasets (which you may use in place of US PPs). A research group generated PPs (not yet tested though) for the Lanthanides series and they kindly agreed to make them public. See http://www.democritos.it/pipermail/pw_forum/2009-May/012897.html GS > > Thanks a lot > > > Wen Shen > Ph. D candidate > Physics department > Georgetown University > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From sclauzer at sissa.it Wed May 27 08:55:46 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Wed, 27 May 2009 08:55:46 +0200 Subject: [Pw_forum] Molybdenum pseudopotentials In-Reply-To: <9d6ea92e0905251040t464c378di993000dd3b472d0b@mail.gmail.com> References: <9d6ea92e0905251040t464c378di993000dd3b472d0b@mail.gmail.com> Message-ID: <4A1CE3F2.7060201@sissa.it> Dear Isaac (from??? please provide your affiliation, thanks) there is a PW91 Mo PP in the QE website. I think you could try to generate a PBE version starting from the same cut-off radii. To my knowledge, PW91 and PBE are not much different, so I guess that starting from the PW91 parameters to generate a PBE could work. Gabriele isaac motochi wrote: > Dear friends, > > Please can anyone sent me PBE pseudopotentials for molybdenum. > > Regards > > Isaac > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From giannozz at democritos.it Wed May 27 09:36:03 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Wed, 27 May 2009 09:36:03 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro In-Reply-To: <4A1C15B1.5010902@unibo.it> References: <4A1C15B1.5010902@unibo.it> Message-ID: <46AD817B-ED4B-4785-A9CF-244147A802EC@democritos.it> On May 26, 2009, at 18:15 , Cristian Degli Esposti Boschi wrote: > on the Mac Pro I get [...] > Operating system error: Cannot allocate memory I never had problems with the gfortran for Mac you can find here: http://gcc.gnu.org/wiki/GFortran. P. --- Paolo Giannozzi, Democritos and University of Udine, Italy From sclauzer at sissa.it Wed May 27 09:56:35 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Wed, 27 May 2009 09:56:35 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro In-Reply-To: <4A1C15B1.5010902@unibo.it> References: <4A1C15B1.5010902@unibo.it> Message-ID: <4A1CF233.7010705@sissa.it> Dear Cristian Cristian Degli Esposti Boschi wrote: > > Ok, in order to simplify the matter I restrict myself to a serial build. > Therefore I pass the option "--disable-parallel" to configure and make > all (using gfortran). I tried both pw.x and cp.x with two input files > for a rather demanding system; on a 2 Xeon Quad-core machine with 16 GB > RAM very similar to the Mac Pro they are successfully processed while > on the Mac Pro I get: > > One thing I noticed in the file config.log is that there > is no malloc usability. The file malloc.h is not found, so > I made a symlink from /usr/include/malloc/malloc.h to > /usr/include/malloc.h. But then the same problem happens > with "struct mallinfo.arena"... As you pointed this out I checked my config.log and I see the same warning when compiling with ifort/icc, but it seems to make no harm at the end. Maybe the problem is in the GCC compiler (I suppose you're using gcc, am I right?). Which version are you using? I cannot afford very demanding calculation, since I've got Leopard on my laptop only. So far I've tried a pw.x calculation which asks up to about 1GB or RAM and it seems to work (for sure it does not crash). > > Is there a special variable to pass to locate correctly > the (malloc related) include files under OSX Leopard 10.5, maybe when I > use the compilers from FINK? > > > Finally, what I meant about 64-bit compiling is that I used to get > similar errors with a C++ code of mine: So the problem might be in the c++ compiler, not the f90 one, as I was saying above (just a guess). Can you recompile with another c compiler/version? GS > > dmrgcc(45305) malloc: *** mmap(size=487424) failed (error code=12) > *** error: can't allocate region > *** set a breakpoint in malloc_error_break to debug > terminate called after throwing an instance of 'std::bad_alloc' > what(): St9bad_alloc > > In that case I solved the problem by passing explicitely the > option "-m64" to the compilers; with Quantum Espresso the configure step > with -m64 is ok, but when I make all I get several errors in fft_scalar. > > > Thanks for your time. Cristian -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From akohlmey at cmm.chem.upenn.edu Wed May 27 13:27:34 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 27 May 2009 07:27:34 -0400 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro In-Reply-To: <4A1C15B1.5010902@unibo.it> References: <4A1C15B1.5010902@unibo.it> Message-ID: <7b6913e90905270427g20164b47g99527c1a6213ec64@mail.gmail.com> On Tue, May 26, 2009 at 12:15 PM, Cristian Degli Esposti Boschi wrote: > Therefore I pass the option "--disable-parallel" to configure and make > all (using gfortran). I tried both pw.x and cp.x with two input files > for a rather demanding system; on a 2 Xeon Quad-core machine with 16 GB > RAM very similar to the Mac Pro they are successfully processed while > on the Mac Pro I get: > > > pw.x(45534) malloc: *** mmap(size=1093816320) failed (error code=12) > *** error: can't allocate region > *** set a breakpoint in malloc_error_break to debug > Operating system error: Cannot allocate memory > Out of memory well, your operating system tells you that you are running out of memory. this is not the fault of Q-E. perhaps the problem is too big for the machine? > or > > cp.x(45549) malloc: *** mmap(size=546910208) failed (error code=12) > *** error: can't allocate region > *** set a breakpoint in malloc_error_break to debug > Operating system error: Cannot allocate memory > Out of memory > > > One thing I noticed in the file config.log is that there > is no malloc usability. The file malloc.h is not found, so > I made a symlink from /usr/include/malloc/malloc.h to > /usr/include/malloc.h. But then the same problem happens > with "struct mallinfo.arena"... this is nonsense. malloc.h is an optional header. it does not have to be there. never mess with system headers. all hell can break lose. > > Is there a special variable to pass to locate correctly > the (malloc related) include files under OSX Leopard 10.5, maybe when I > use the compilers from FINK? > > > Finally, what I meant about 64-bit compiling is that I used to get > similar errors with a C++ code of mine: > > dmrgcc(45305) malloc: *** mmap(size=487424) failed (error code=12) > *** error: can't allocate region > *** set a breakpoint in malloc_error_break to debug > terminate called after throwing an instance of 'std::bad_alloc' > ? what(): ?St9bad_alloc well, same thing. you are running out of memory. > In that case I solved the problem by passing explicitely the > option "-m64" to the compilers; with Quantum Espresso the configure step > with -m64 is ok, but when I make all I get several errors in fft_scalar. ok. now that explains something. your default compiler installation is a 32-bit version. those have a limited address space of about 2GB. this is why you are running out of memory. when you change to 64-bit mode, this is resolved. however, you cannot just change make.sys. you have to re-run configure and pass the -m64 flag with the compiler flags (for c, f77 _and_ f90) and then compilation should work. in 64-bit mode some parts of the fortran and c code in Q-E have to be adjusted due to the difference in pointer size vs. the default integer size compared to 32-bit mode. cheers,\ axel. > > Thanks for your time. Cristian > -- > ___________________________________________________ > Cristian Degli Esposti Boschi > CNR, CNISM, Unita' di Ricerca di Bologna, c/o > Dipartimento di Fisica, Universita' di Bologna > viale Berti-Pichat, 6/2, 40127, Bologna, Italia > tel. ++39 051 2095114 ? fax ++39 051 2095113 > e-mail: cristian.degliesposti -AT- unibo.it > web: ? ?http://www.df.unibo.it/fismat/theory > ___________________________________________________ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From pc229 at kent.ac.uk Wed May 27 16:39:44 2009 From: pc229 at kent.ac.uk (Pieremanuele Canepa) Date: Wed, 27 May 2009 15:39:44 +0100 Subject: [Pw_forum] input for resp_mat.x in to optimize the U for LDA+U calculations Message-ID: <8b443ba10905270739w463d2b36med5b040b59c1050@mail.gmail.com> Dear all, I was able to compile the program resp_mat.f90 in order to carry out the U in a self consistent way. Now I am trying to write up correctly the input for resp_mat.x. I have realized reading trough the guide (found at link http://vlab.msi.umn.edu/events/lecture.shtml) that I do need to have an external file with the fractional coordinate of the atom belonging to the supercell. Actually, if I look at those external file (call for example pos....) that come togheter with the guide found online, the second part is concerning to the fractional coordinate of the atoms in the supercell, while what is the first part? Are these the vector of the supercell in Cartesian coordinates ? Why, when for instance the Cartesian coordinate of the supercell vectors printed in the early part of the PWscf output are: crystal axes: (cart. coord. in units of a_0) a(1) = ( 1.000000 0.000000 0.000000 ) a(2) = ( 0.000000 1.000000 0.000000 ) a(3) = ( 0.000000 0.000000 1.000000 ) in the file pos these become 20.d0 0.d0 0.d0 0.d0 20.d0 0.d0 0.d0 0.d0 20.d0 ? The cell parameter in this case is 10.8400 (FeNi in the guide) Another question is : what does the line back = 'neutral' mean in the input file for resp_mat.x? I hope you can get my point! Best Regards, Piero -- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom ----------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090527/9aef8ec8/attachment.htm From swblelia at ehu.es Wed May 27 20:44:38 2009 From: swblelia at ehu.es (Aritz Leonardo Liceranzu) Date: Wed, 27 May 2009 20:44:38 +0200 Subject: [Pw_forum] inhomogeneous K-point sampling Message-ID: <20090527204438.bruclmacogogg4o0@www.ehu.es> Hi all, I wrote a little program that computes the binding energies of excitons for different semiconductor systems like GaAs, GaN CdSe ...etc. Starting from equilibrium KS DFT wave-functions and making use of linear response theory exciton binding energies maybe calculated. Basically one has to digonalize the momentum dependent exchange-correlation kernel F_kq(r,r',w). To test the convergence of my results in terms of the k-point sampling I would need to increase the number of K,q points considered to build the above matrix. Considering that the code I wrote is serial and is far from being computationally optimized the number of k points considered makes a big deal in terms of computation time. On the other hand I have the feeling that for the description of excitons it is important to consider "many" kpoints in the surroundings of the Gamma point where the valence band max. and the conduction bands minimum are. But no so many far from Gamma. Of course, I could always use a huge k grid but that is computationally impossible for me. So, the question would be, is it stupid to do the following? Take a huge grid, let us say (14,14,14) and get rid of points that are at farther than a certain threshold from Gamma, normalizing the weights of the "new" grid so that they sum one. Thank you very much for your help and feel free to say that it is a stupid idea. Aritz From baroni at sissa.it Wed May 27 21:23:23 2009 From: baroni at sissa.it (Stefano Baroni) Date: Wed, 27 May 2009 21:23:23 +0200 Subject: [Pw_forum] inhomogeneous K-point sampling In-Reply-To: <20090527204438.bruclmacogogg4o0@www.ehu.es> References: <20090527204438.bruclmacogogg4o0@www.ehu.es> Message-ID: <8F163890-6B08-432D-974F-00B5C1A296A1@sissa.it> On May 27, 2009, at 8:44 PM, Aritz Leonardo Liceranzu wrote: > Hi all, Hi Leonardo: > So, the question would be, is it stupid to do the following? > > Take a huge grid, let us say (14,14,14) and get rid of points that are > at farther than a certain threshold from Gamma, normalizing the > weights of the "new" grid so that they sum one. This not stupid, of course. It may be wrong, though. There are many different ways of "normalizing the new wights of the new grid so that they sum to one". Some of them correct (within a given accuracy), others plain wrong. It all depends on how you choose the weights and on the specific integral you aim at approximating with a weighted sum. Ever read a textbook on numerical integration? This and similar issues are treated, I think, in many textbooks on numerical analysis. Please, take a glance at one of them and then revert to us. S. --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090527/2ab67b07/attachment-0001.htm From ismotochi at gmail.com Wed May 27 21:47:02 2009 From: ismotochi at gmail.com (isaac motochi) Date: Wed, 27 May 2009 07:47:02 -1200 Subject: [Pw_forum] Molybdenum pseudopotentials In-Reply-To: <4A1CE3F2.7060201@sissa.it> References: <9d6ea92e0905251040t464c378di993000dd3b472d0b@mail.gmail.com> <4A1CE3F2.7060201@sissa.it> Message-ID: <9d6ea92e0905271247s5085605bq7bb8d4b73a41f77c@mail.gmail.com> Thanks Gabriele, I ll try the pp and keep you informed. Cheers, Motochi Isaac Moi University, Kenya On 5/26/09, Gabriele Sclauzero wrote: > Dear Isaac (from??? please provide your affiliation, thanks) > > there is a PW91 Mo PP in the QE website. I think you could try to > generate a PBE > version starting from the same cut-off radii. To my knowledge, PW91 and PBE > are not much > different, so I guess that starting from the PW91 parameters to generate a > PBE could work. > > Gabriele > > isaac motochi wrote: >> Dear friends, >> >> Please can anyone sent me PBE pseudopotentials for molybdenum. >> >> Regards >> >> Isaac >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > -- > > > o ------------------------------------------------ o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > | via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511 | > | skype: gurlonotturno | > o ------------------------------------------------ o > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From swblelia at ehu.es Thu May 28 06:00:33 2009 From: swblelia at ehu.es (Aritz Leonardo Liceranzu) Date: Wed, 27 May 2009 23:00:33 -0500 Subject: [Pw_forum] inhomogeneous K-point sampling In-Reply-To: <8F163890-6B08-432D-974F-00B5C1A296A1@sissa.it> References: <20090527204438.bruclmacogogg4o0@www.ehu.es> <8F163890-6B08-432D-974F-00B5C1A296A1@sissa.it> Message-ID: Dear prof. Baroni, Thanks for your prompt answer. "Ever read a textbook on numerical integration? This and similar issues are treated, I think, in many textbooks on numerical analysis." Up to know the integrals I calculated were performed in real space using Simpson's rule and was enough. To build the (k,q) momentum dependent F_kq(r,r',w) matrix, I have to pick with caution wavefunctions and their corresponding weights. It is not that I have to integrate them but build a matrix and diagonalize it. As I said in my 1st email choosing a uniform small K-grid to build F does not assure me convergence, that is why I would like to add points close to the Gamma point. For this purpose, I need to be able to construct nonuniform reciprocal grids in the IBZ with their corresponding wafunctions and weights. I am trying to figure out how to do this so I guess that first I will need to understand how Monkhorst-Pack grids are generated and more precisely the corresponding the weights of each kpoint. ANY REFERENCE AND/OR TEXTBOOK IS REALLY APPRECIATED. (I just printed the original M-P paper so I can read it... ) Thanks again Aritz On 09-mai-27, at 14:23, Stefano Baroni wrote: > > On May 27, 2009, at 8:44 PM, Aritz Leonardo Liceranzu wrote: > >> Hi all, > > Hi Leonardo: > >> So, the question would be, is it stupid to do the following? >> >> Take a huge grid, let us say (14,14,14) and get rid of points that >> are >> at farther than a certain threshold from Gamma, normalizing the >> weights of the "new" grid so that they sum one. > > This not stupid, of course. It may be wrong, though. There are many > different ways of "normalizing the new wights of the new grid so > that they sum to one". Some of them correct (within a given > accuracy), others plain wrong. It all depends on how you choose the > weights and on the specific integral you aim at approximating with > a weighted sum. Ever read a textbook on numerical integration? This > and similar issues are treated, I think, in many textbooks on > numerical analysis. Please, take a glance at one of them and then > revert to us. > > S. > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - > Trieste > http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / > stefanobaroni (skype) > > La morale est une logique de l'action comme la logique est une > morale de la pens?e - Jean Piaget > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090527/441cf981/attachment.htm From baroni at sissa.it Thu May 28 08:17:29 2009 From: baroni at sissa.it (Stefano Baroni) Date: Thu, 28 May 2009 08:17:29 +0200 Subject: [Pw_forum] inhomogeneous K-point sampling In-Reply-To: References: <20090527204438.bruclmacogogg4o0@www.ehu.es> <8F163890-6B08-432D-974F-00B5C1A296A1@sissa.it> Message-ID: On May 28, 2009, at 6:00 AM, Aritz Leonardo Liceranzu wrote: > Dear prof. Baroni, Aritz: > For this purpose, I need to be able to construct nonuniform > reciprocal grids in the IBZ with their corresponding wafunctions and > weights. I am trying to figure out how to do this so I guess that > first I will need to understand how Monkhorst-Pack grids are > generated and more precisely the corresponding the weights of each > kpoint. MP grids are uniform grids in reciprocal space. Doing Simpson integration on a regular grid would just require equal weigths. The reason why weights are not equal is because one does not use all the points of the grid, but just those that are inequivelent by symmetry (i.e. that belong to the "irreducible wedge" of the BZ). The weight in this case is just equal to the total number of points that are equivalent to a given point ("the star" of a given point). An example will explain this. Suppose you are in cubic symmetry and that you have the points (xyz), (xy0), (x00). What are the corresponding weights? Well, there are 48 points equivalent to (xyz), 24 points equivalent to (xy0) and 6 equivalent to (x00). The relative wights will thus be 8, 4, 1. A special care must be paid to points that lie on the border of the BZ. In this case, a symmetry operation may send a point into itself (modulo a reciprocal-lattice vector). For instance, the star of (1/2,0,0) has six elements for an FCC lattice, whereas it has only 3 for a SC lattice [because (1/2,0,0) and (-1/2,0,0) differ by (100) which is a reciprocal-lattice vector for SC but not for FCC]. SO far, so good, but not easily generalizable to non-uniform grids. The generalization requires a slight shift of point of view. The weight can be seen as the measure of a small volume associated to each integration points. Points in the interior of the irreducible wedge all have the same weight, because the grid is uniform. Points on the surface of the IW will have different weights according to the way the surface cuts the elementary volumes of the uniform grids. If you are good at geometry (I am not) you can recover the same weights that you get from symmetry considerations. This way of proceeding has the advantage of being easily generalizable (at least in principle) to non- uniform grids. You have to devise a way to tile the entire space with small volumes, each of them associated with each point of your grid. The "weight" of the point will the be just the volume of the 3D tile associated to it. > ANY REFERENCE AND/OR TEXTBOOK IS REALLY APPRECIATED. (I just printed > the original M-P paper so I can read it... ) Non-uniform grids are treated in texts on the numerical solution of partial differential equations (search for "finite differences" or "finite elements"). I think that the most straightorward way of tiling a 3D volume is by using tetrahedra, wich are 3D simplexes. A simplex is the nD polyhedron delimited by the smallest number of (n-1)D surfaces. 2D simplexes are triangles, 3D simplexes are tetrahedra. I think that there automatic ways of generating tetrahedron tiles, but I do not know exactly. If I were you, I would try to figure out by myself. If you fail after 1-2 days of hard trying, then a library (or, may be, google) could come to the rescue. Hope this helps. Coraggio! Stefano --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) La morale est une logique de l'action comme la logique est une morale de la pens?e - Jean Piaget Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090528/c9f9710d/attachment-0001.htm From anavat_ila at yahoo.com Thu May 28 09:48:29 2009 From: anavat_ila at yahoo.com (Ali Tavana) Date: Thu, 28 May 2009 00:48:29 -0700 (PDT) Subject: [Pw_forum] LSDA+U bandstructure Message-ID: <672829.90610.qm@web51812.mail.re2.yahoo.com> Dear friends, I have a question about LSDA+U calculations. I have done a LSDA+U and a LDA calculation for a system, but surprisingly there is no difference between bandstructures obtained within them. LSDA+U has found a nonmagnetic answer for my compound (my system is magnetic in fact), but I should see some differences between LDA and LSDA+U as the Hubbard U should separate spin-up and spin-down electron bands. Can one explain for me what is there going on? Ali Atomistic Modeling and Design of Materials, University of Leoben, Austria From anavat_ila at yahoo.com Thu May 28 10:00:51 2009 From: anavat_ila at yahoo.com (Ali Tavana) Date: Thu, 28 May 2009 01:00:51 -0700 (PDT) Subject: [Pw_forum] LSDA+U phonons Message-ID: <903650.105.qm@web51804.mail.re2.yahoo.com> Dear friends, Does any body know whether phonon calculations can be carried out for spin-LDA (LSDA) or not? It seems that it is not implemented in QE or maybe its implemented but some one should do something unusual to use it ;)? (I saw papers by Corso, about spin polarized calculation of phonons with QE) From sclauzer at sissa.it Thu May 28 10:07:58 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 28 May 2009 10:07:58 +0200 Subject: [Pw_forum] LSDA+U phonons In-Reply-To: <903650.105.qm@web51804.mail.re2.yahoo.com> References: <903650.105.qm@web51804.mail.re2.yahoo.com> Message-ID: <4A1E465E.4080802@sissa.it> Ali Tavana wrote: > Dear friends, > > Does any body know whether phonon calculations can be carried out for spin-LDA (LSDA) > or not? It seems that it is not implemented in QE or maybe its implemented but some one > should do something unusual to use it ;)? (I saw papers by Corso, about spin polarized > calculation of phonons with QE) I think that the LSDA phonon calculation can be carried out without any need to do "something unusual". BTW, which paper by Dal Corso are you refering to? If you meant phonos within LSDA+U, they have not been implemented. GS > > > > _______________________________________________ Pw_forum mailing list > Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From pc229 at kent.ac.uk Thu May 28 10:08:25 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Thu, 28 May 2009 09:08:25 +0100 Subject: [Pw_forum] input for resp_mat.x in to optimize the U for LDA+U calculations Message-ID: Dear all, I was? able to compile? the program resp_mat.f90 in order to? carry out the U in a self consistent way. Now I am trying to write up correctly the input for resp_mat.x. I have realized reading trough the guide (found at link http://vlab.msi.umn.edu/events/lecture.shtml)? that I do need to have an external file?? with the fractional coordinate of the atom belonging to the supercell. Actually, if I look at those external file (call for example pos....) that come togheter with the guide found online, the second part is concerning to the fractional coordinate of the atoms in the supercell, while what is the first part? Are these the vector of the supercell in Cartesian coordinates ? Why, when? for instance? the ? Cartesian coordinate of the? supercell vectors printed in the early part of the PWscf output are: ? crystal axes: (cart. coord. in units of a_0) ?????????????? a(1) = (? 1.000000? 0.000000? 0.000000 )? ?????????????? a(2) = (? 0.000000? 1.000000? 0.000000 )? ?????????????? a(3) = (? 0.000000? 0.000000? 1.000000 )? in the file pos? these become 20.d0 0.d0 0.d0 0.d0 20.d0 0.d0 0.d0 0.d0 20.d0 ? The cell parameter? in this case is 10.8400 (FeNi in the guide) Another question is : what does the line ? back = 'neutral' mean? in the input file for resp_mat.x? I hope? you can get my point! Best Regards, Piero --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090528/16d8a169/attachment.htm From sclauzer at sissa.it Thu May 28 10:14:51 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 28 May 2009 10:14:51 +0200 Subject: [Pw_forum] LSDA+U bandstructure In-Reply-To: <672829.90610.qm@web51812.mail.re2.yahoo.com> References: <672829.90610.qm@web51812.mail.re2.yahoo.com> Message-ID: <4A1E47FB.3090702@sissa.it> Ali Tavana wrote: > Dear friends, > > I have a question about LSDA+U calculations. I have done a LSDA+U and a LDA calculation > for a system, but surprisingly there is no difference between bandstructures obtained > within them. That's strange, unless there is something wrong in your setup or you use a very small U, you should see the position of "Hubbard electrons" bands move. Which material are you studying? > LSDA+U has found a nonmagnetic answer for my compound (my system is > magnetic in fact), but I should see some differences between LDA and LSDA+U as the > Hubbard U should separate spin-up and spin-down electron bands. Can one explain for me > what is there going on? In order to get a magnetic GS you have to start from a magnetic configuration. Have you? Then you can see if the system loses or not the magnetization (with U) and, if not, compare the energies of the magnetic and non-magnetic solutions (both with U). GS > > Ali Atomistic Modeling and Design of Materials, University of Leoben, Austria > > > > _______________________________________________ Pw_forum mailing list > Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum > -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From anavat_ila at yahoo.com Thu May 28 10:35:54 2009 From: anavat_ila at yahoo.com (Ali Tavana) Date: Thu, 28 May 2009 01:35:54 -0700 (PDT) Subject: [Pw_forum] Pw_forum Digest, Vol 23, Issue 64 Message-ID: <157014.67730.qm@web51803.mail.re2.yahoo.com> Dear Gabriele, Yes, I had a starting magnetization but it suppressed after some iterations. The U is about 6 eV, and my compound is a high Tc cuprate. The difference between non-spin and spin input files is: nspin=2, starting_magnetization(1)=1 Hubbard_U(1)=5.578 And about the phonons, I know one can calculate the phonons with LSDA in QE but in fact I was meaning the e-ph interaction, which seems is not implemented yet neither within LSDA or LSDA+U methods. From matteo at umn.edu Thu May 28 11:09:17 2009 From: matteo at umn.edu (matteo at umn.edu) Date: 28 May 2009 04:09:17 -0500 Subject: [Pw_forum] input for resp_mat.x in to optimize the U for LDA+U calculations In-Reply-To: <8b443ba10905270739w463d2b36med5b040b59c1050@mail.gmail.com> References: <8b443ba10905270739w463d2b36med5b040b59c1050@mail.gmail.com> Message-ID: Dear Pieremanuele, if you don't have the crystal coordinates of the atoms you can have the code print them. On May 27 2009, Pieremanuele Canepa wrote: >Dear all, >I was able to compile the program resp_mat.f90 in order to carry out the >U in a self consistent way. Now I am trying to write up correctly the input >for resp_mat.x. I have realized reading trough the guide (found at link >http://vlab.msi.umn.edu/events/lecture.shtml) that I do need to have an >external file with the fractional coordinate of the atom belonging to the >supercell. >Actually, if I look at those external file (call for example pos....) that >come togheter with the guide found online, the second part is concerning to >the fractional coordinate of the atoms in the supercell, while what is the > first part? Are these the vector of the supercell in Cartesian > coordinates ? yes. >Why, when for instance the Cartesian coordinate of the supercell >vectors printed in the early part of the PWscf output are: > crystal axes: (cart. coord. in units of a_0) > a(1) = ( 1.000000 0.000000 0.000000 ) > a(2) = ( 0.000000 1.000000 0.000000 ) > a(3) = ( 0.000000 0.000000 1.000000 ) >in the file pos these become >20.d0 0.d0 0.d0 >0.d0 20.d0 0.d0 >0.d0 0.d0 20.d0 >? it doesn't matter the length of the vectors. only the ratio between length does. This is used to compute the distance between atoms ut the scale doesn't matter at all. >The cell parameter in this case is 10.8400 (FeNi in the guide) > >Another question is : what does the line back = 'neutral' mean in the >input file for resp_mat.x? > neutral means that you are enforcing the total response of the system to be neutral. so there will be a compensating "background" term in the response matrix that results in an additional column and row. see PRB 71 35105. hope this helps. regards, Matteo I hope you can get my point! >Best Regards, Piero > From pc229 at kent.ac.uk Thu May 28 11:28:18 2009 From: pc229 at kent.ac.uk (pc229 at kent.ac.uk) Date: Thu, 28 May 2009 10:28:18 +0100 Subject: [Pw_forum] input for resp_mat.x in to optimize the U for LDA+U calculations In-Reply-To: References: <8b443ba10905270739w463d2b36med5b040b59c1050@mail.gmail.com> Message-ID: Thanks a lot Matteo! ----- Original Message ----- From: matteo at umn.edu Date: Thursday, May 28, 2009 10:24 am Subject: Re: [Pw_forum] input for resp_mat.x in to optimize the U for LDA+U calculations To: PWSCF Forum > > Dear Pieremanuele, > > if you don't have the crystal coordinates of the atoms you can > have the > code print them. > > > On May 27 2009, Pieremanuele Canepa wrote: > > >Dear all, > >I was? able to compile? the program resp_mat.f90 in > order to? carry out the > >U in a self consistent way. Now I am trying to write up > correctly the input > >for resp_mat.x. I have realized reading trough the guide (found > at link > >http://vlab.msi.umn.edu/events/lecture.shtml)? that I do > need to have an > >external file?? with the fractional coordinate of the > atom belonging to the > >supercell. > >Actually, if I look at those external file (call for example > pos....) that > >come togheter with the guide found online, the second part is > concerning to > >the fractional coordinate of the atoms in the supercell, while > what is the > > first part? Are these the vector of the supercell in Cartesian > > coordinates ? > > yes. > > >Why, when? for instance? the?? Cartesian > coordinate of the? supercell > >vectors printed in the early part of the PWscf output are: > >? crystal axes: (cart. coord. in units of a_0) > >?????????????? a(1) = (? 1.000000? 0.000000? 0.000000 ) > >?????????????? a(2) = (? 0.000000? 1.000000? 0.000000 ) > >?????????????? a(3) = (? 0.000000? 0.000000? 1.000000 ) > >in the file pos? these become > >20.d0 0.d0 0.d0 > >0.d0 20.d0 0.d0 > >0.d0 0.d0 20.d0 > >? > > it doesn't matter the length of the vectors. only the ratio > between length > does. > This is used to compute the distance between atoms ut the scale > doesn't > matter at all. > > > >The cell parameter? in this case is 10.8400 (FeNi in the guide) > > > >Another question is : what does the line?? back = > 'neutral' mean? in the > >input file for resp_mat.x? > > > > neutral means that you are enforcing the total response of the > system to be > neutral. so there will be a compensating "background" term in > the response > matrix that results in an additional column and row. see PRB 71 35105. > > hope this helps. > > regards, > > Matteo > > > I hope? you can get my point! > >Best Regards, Piero > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Pieremanuele Canepa Room 230 School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent, CT2 7NH United Kingdom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090528/a98cb287/attachment-0001.htm From cristian.degliesposti at unibo.it Thu May 28 12:12:40 2009 From: cristian.degliesposti at unibo.it (Cristian Degli Esposti Boschi) Date: Thu, 28 May 2009 12:12:40 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro Message-ID: <4A1E6398.9060307@unibo.it> Dear all, thanks for the replies. Here a brief list of comments to each one of them: >I never had problems with the gfortran for Mac you can find here: >http://gcc.gnu.org/wiki/GFortran. In this page I've seen that Gfortran is known to work on powerpc-darwin and i386-darwin. My architecture is i386-apple-darwin9.0. In fact I also never had problems before with gfortran on that Mac. >As you pointed this out I checked my config.log and I see the same >warning when compiling >with ifort/icc, but it seems to make no harm at the end. Maybe the >problem is in the GCC >compiler (I suppose you're using gcc, am I right?). Which version are >you using? gcc version 4.0.1 (Apple Inc. build 5490) [others available 4.3.3 (GCC) and gcc 4.2.1 (Apple Inc. build 5566)] >well, your operating system tells you that you are running out >of memory. this is not the fault of Q-E. >perhaps the problem is too big for the machine? The same problem runs without crashing on another machine without OSX but with the same processors and amount of RAM... >ok. now that explains something. your default compiler installation >is a 32-bit version. those have a limited address space of about >2GB. this is why you are running out of memory. when you change >to 64-bit mode, this is resolved. however, you cannot just change >make.sys. you have to re-run configure and pass the -m64 flag >with the compiler flags (for c, f77 _and_ f90) and then compilation >should work. I've passed the -m64 option at the very beginning in the configure step but then when I make all I get the error and warning with fft_scalar ! >in 64-bit mode some parts of the fortran and c code >in Q-E have to be adjusted due to the difference in pointer size >vs. the default integer size compared to 32-bit mode. So, do I have to adjust the code? Where and how? Not like that in the other 64-bit machine without OSX... THANKS again for your time. Cristian -- ___________________________________________________ Cristian Degli Esposti Boschi CNR, CNISM, Unita' di Ricerca di Bologna, c/o Dipartimento di Fisica, Universita' di Bologna viale Berti-Pichat, 6/2, 40127, Bologna, Italia tel. ++39 051 2095114 fax ++39 051 2095113 e-mail: cristian.degliesposti -AT- unibo.it web: http://www.df.unibo.it/fismat/theory ___________________________________________________ From sclauzer at sissa.it Thu May 28 12:29:17 2009 From: sclauzer at sissa.it (Gabriele Sclauzero) Date: Thu, 28 May 2009 12:29:17 +0200 Subject: [Pw_forum] Installing QE 4.0.5 on Mac Pro In-Reply-To: <4A1E6398.9060307@unibo.it> References: <4A1E6398.9060307@unibo.it> Message-ID: <4A1E677D.1010003@sissa.it> Cristian Degli Esposti Boschi wrote: > >As you pointed this out I checked my config.log and I see the same > >warning when compiling > >with ifort/icc, but it seems to make no harm at the end. Maybe the > >problem is in the GCC > >compiler (I suppose you're using gcc, am I right?). Which version are > >you using? > > > gcc version 4.0.1 (Apple Inc. build 5490) > [others available 4.3.3 (GCC) and gcc 4.2.1 (Apple Inc. build 5566)] Have you tried with 4.3.x for instance (not Apple build)? There should be also 4.4.x available. Anyway, if you really want to do big calculations on a Mac with Intel CPU, I strongly suggest to use ifort/icc (for simple tests on pw.x I found that the speed-up can be huge). It is not free for Mac, but you can buy student/academic licenses (I bought a student suite for 40 euros, with ifort, icc, mkl, idb, ipp, ... all included). GS -- o ------------------------------------------------ o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | | via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o ------------------------------------------------ o From giuseppe.mattioli at mlib.ism.cnr.it Thu May 28 12:43:16 2009 From: giuseppe.mattioli at mlib.ism.cnr.it (Giuseppe Mattioli) Date: Thu, 28 May 2009 12:43:16 +0200 Subject: [Pw_forum] LSDA+U bandstructure In-Reply-To: <157014.67730.qm@web51803.mail.re2.yahoo.com> References: <157014.67730.qm@web51803.mail.re2.yahoo.com> Message-ID: <200905281243.17121.giuseppe.mattioli@mlib.ism.cnr.it> have you switched on the U correction? lda_plus_u=.true. I can understand that you don't want to show the whole of your input file, but such a choice makes useful suggestions very unlikely to come... Giuseppe On Thursday 28 May 2009 10:35:54 Ali Tavana wrote: > Dear Gabriele, > > Yes, I had a starting magnetization but it suppressed after some > iterations. The U is about 6 eV, and my compound is a high Tc cuprate. > > The difference between non-spin and spin input files is: > > nspin=2, > starting_magnetization(1)=1 > Hubbard_U(1)=5.578 > > And about the phonons, I know one can calculate the phonons with LSDA in QE > but in fact I was meaning the e-ph interaction, which seems is not > implemented yet neither within LSDA or LSDA+U methods. > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ******************************************************** - Article premier - Les hommes naissent et demeurent libres et ?gaux en droits. Les distinctions sociales ne peuvent ?tre fond?es que sur l'utilit? commune - Article 2 - Le but de toute association politique est la conservation des droits naturels et imprescriptibles de l'homme. Ces droits sont la libert?, la propri?t?, la s?ret? et la r?sistance ? l'oppression. ******************************************************** Giuseppe Mattioli CNR - ISTITUTO DI STRUTTURA DELLA MATERIA v. Salaria Km 29,300 - C.P. 10 I 00016 - Monterotondo Stazione (RM) Tel + 39 06 90672836 - Fax +39 06 90672316 E-mail: -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090528/0f7a6042/attachment.htm From marzari at MIT.EDU Thu May 28 12:47:19 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 28 May 2009 06:47:19 -0400 Subject: [Pw_forum] inhomogeneous K-point sampling In-Reply-To: References: <20090527204438.bruclmacogogg4o0@www.ehu.es> <8F163890-6B08-432D-974F-00B5C1A296A1@sissa.it> Message-ID: <4A1E6BB7.3010707@mit.edu> > On May 28, 2009, at 6:00 AM, Aritz Leonardo Liceranzu wrote: >> For this purpose, I need to be able to construct nonuniform reciprocal >> grids in the IBZ with their corresponding wafunctions and weights. I >> am trying to figure out how to do this so I guess that first I will >> need to understand how Monkhorst-Pack grids are generated and more >> precisely the corresponding the weights of each kpoint. Dear Aritz, would Wannier interpolation work for this ? I.e. you calculate Bloch states on a Monkhorst-Pack mesh, transform them into (maximally-localized) Wannier functions, and then interpolate back at any arbitrary k-point ? There is a detailed 2007 PRB on this, by Yates/Souza/Vanderbilt. nicola -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From anavat_ila at yahoo.com Thu May 28 15:03:05 2009 From: anavat_ila at yahoo.com (Ali Tavana) Date: Thu, 28 May 2009 06:03:05 -0700 (PDT) Subject: [Pw_forum] (no subject) Message-ID: <528099.78571.qm@web51810.mail.re2.yahoo.com> thank you very much Giuseppe. I am very sorry, I had forgotten it. From sunyang198498 at gmail.com Thu May 28 21:31:16 2009 From: sunyang198498 at gmail.com (=?GB2312?B?y+/R9A==?=) Date: Thu, 28 May 2009 14:31:16 -0500 Subject: [Pw_forum] Fermi energy Message-ID: <98e7a0810905281231g5c493f49j702206d17cbe236a@mail.gmail.com> Dearl all pw users: I have a problem with the estimated Fermi energy in the scf and nscf calculation toward a slab structure metal. I found that the scf gives a much higher Fermi energy than the one in the nscf calculation. Anybody has idea about this? Best regards, Yang -- University of Nebraska at Lincoln From paulatto at sissa.it Thu May 28 21:35:08 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 28 May 2009 21:35:08 +0200 (CEST) Subject: [Pw_forum] Fermi energy In-Reply-To: <98e7a0810905281231g5c493f49j702206d17cbe236a@mail.gmail.com> References: <98e7a0810905281231g5c493f49j702206d17cbe236a@mail.gmail.com> Message-ID: <50248.78.12.164.220.1243539308.squirrel@webmail.sissa.it> On Thu, May 28, 2009 21:31, ???? wrote: > Anybody has idea about this? The Fermi energy can be a bit different between a scf and a corresponding nscf calculation. If they are very different there is probably an error in your input file (different atomic coordinates, wrong number of electrons...) Weithout seeing you input its difficult to tell, though. Try and compare the inputs and the outputs of your calculations using the program "diff". regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From ferretti at MIT.EDU Thu May 28 21:44:45 2009 From: ferretti at MIT.EDU (Andrea Ferretti) Date: Thu, 28 May 2009 15:44:45 -0400 (EDT) Subject: [Pw_forum] Fermi energy In-Reply-To: <98e7a0810905281231g5c493f49j702206d17cbe236a@mail.gmail.com> References: <98e7a0810905281231g5c493f49j702206d17cbe236a@mail.gmail.com> Message-ID: Dear Yang, The computed fermi energy depends on the adopted kpt grid. Fermi energy is computed in scf runs (assuming occupations are not "fixed") and re-computed in nscf runs and differences can arise due to differences in the scf and nscf kpt meshes... Note that nscf calculations are meant to deal with summations over the Brillouin zone, which means that kpts and their weight must be chosen accordingly. if instead you want eg to compute quantities at specific kpoints (a typical example is the band structure), your kpt summations might lose a lot of accuracy and the estimation of the fermi energy along with that (in this case use calculation="bands" instead) hope it helps andrea > > I have a problem with the estimated Fermi energy in the scf and nscf > calculation toward > a slab structure metal. I found that the scf gives a much higher Fermi > energy than the one > in the nscf calculation. Anybody has idea about this? > > Best regards, > > Yang > > > -- Andrea Ferretti MIT, Dept Material Science & Engineering bldg 13-4078, 77, Massachusetts Ave, Cambridge, MA Tel: +1 617-452-2455; Skype: andrea_ferretti URL: http://quasiamore.mit.edu Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From laura.henderson1 at mycampus.uoit.ca Thu May 28 22:20:24 2009 From: laura.henderson1 at mycampus.uoit.ca (Laura Henderson) Date: Thu, 28 May 2009 16:20:24 -0400 (EDT) Subject: [Pw_forum] Saving the ion velocities Message-ID: <7110493.1243542024178.JavaMail.laura.henderson1@mycampus.uoit.ca> Hi to all, I have a question about molecular dynamics on Quantum Espresso. We use the Nose thermostat to control the temperature of our system; however, we also need to run free molecular dynamics without the Nose thermostat on. When the thermostat is turned off, the ion velocities reset to zero, so the system is quenched to zero Kelvin. Is there a way to turn off the Nose thermostat without the ion velocities resetting to zero? Thanks Laura --------- Laura Henderson University of Ontario Institute of Technology, Ontario, Canada From yaldachem at gmail.com Fri May 29 02:21:31 2009 From: yaldachem at gmail.com (yaldaa kh) Date: Thu, 28 May 2009 17:21:31 -0700 Subject: [Pw_forum] deguss Message-ID: <47431d9f0905281721n4782c64dw7771d062eb159053@mail.gmail.com> dear all 1) why does the total energy (Escf ) decrease with increase of the deguss value in smearing technique? 2) How can I choose a good value (the best choise) for deguss? thanks a lot yalda student (chemistry) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090528/62f55881/attachment.htm From marzari at MIT.EDU Fri May 29 03:31:26 2009 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 28 May 2009 21:31:26 -0400 Subject: [Pw_forum] deguss In-Reply-To: <47431d9f0905281721n4782c64dw7771d062eb159053@mail.gmail.com> References: <47431d9f0905281721n4782c64dw7771d062eb159053@mail.gmail.com> Message-ID: <4A1F3AEE.5080309@mit.edu> yaldaa kh wrote: > dear all > 1) why does the total energy (Escf ) decrease with increase of the > deguss value in smearing technique? > 2) How can I choose a good value (the best choise) for deguss? > > thanks a lot > yalda > student (chemistry) > 1) It shouldn't for standard smearings (e.g. Fermi-Dirac, gaussian). E-TS decreases. For m-p, m-v, we cannot say. 2) See Chap 4 of http://quasiamore.mit.edu/phd/ for plenty details. nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From degironc at sissa.it Fri May 29 09:03:10 2009 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 29 May 2009 09:03:10 +0200 Subject: [Pw_forum] deguss In-Reply-To: <4A1F3AEE.5080309@mit.edu> References: <47431d9f0905281721n4782c64dw7771d062eb159053@mail.gmail.com> <4A1F3AEE.5080309@mit.edu> Message-ID: <4A1F88AE.1000204@sissa.it> Nicola Marzari wrote: > yaldaa kh wrote: > >> dear all >> 1) why does the total energy (Escf ) decrease with increase of the >> deguss value in smearing technique? >> 2) How can I choose a good value (the best choise) for deguss? >> >> thanks a lot >> yalda >> student (chemistry) >> > > > 1) It shouldn't for standard smearings (e.g. Fermi-Dirac, gaussian). > E-TS decreases. For m-p, m-v, we cannot say. > In PWscf in case of smearing the total energy is the variational quantity and contains an entropy-like term (-TS contibution). In case one is using Fermi-Dirac smearing this variational quantity is the electronic free-energy, for other smearing it is an auxiliary quantity analogue to that but with no relationship with electronic free energy. As Nicola says, for FD and Gaussian smearing "E-TS" (that is total energy in PWscf) decreases with increasing T (smearing). for other smearing type we cannot say. Stefano de Gironcoli - SISSA and DEMOCRITOS > 2) See Chap 4 of http://quasiamore.mit.edu/phd/ for plenty details. > > nicola > > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From rajanpnano at gmail.com Fri May 29 14:05:58 2009 From: rajanpnano at gmail.com (Raj Pandey) Date: Fri, 29 May 2009 17:35:58 +0530 Subject: [Pw_forum] Compilation Error on BlueGene/L Message-ID: <48b8698d0905290505y3fa5455dq6f58852003f5d1bb@mail.gmail.com> Dear PWSCF developers and users, I ran the configure script on BlueGene/L: ./configure MPIF90=mpixlf90 CC=mpixlc LD=mpixlf90 BLAS_LIBS="/bgl/local/lib/libblas440.a" LAPACK_LIBS="/bgl/local/lib/liblapack440.a" FFT_LIBS="/opt/ibmmath/essl/4.2/lib/libesslbg.a" and then "make all", which takes about 30 - 40 minutes and returns with an error. This time is long compared to for example a successful compilation on X86 machines, typically 15-20 minutes on 8 cores blade. The following is the error (tail of the log file): ** xsf_struct === End of Compilation 1 === ** xsf_fast_datagrid_3d === End of Compilation 2 === ** xsf_datagrid_2d === End of Compilation 3 === ** xsf_datagrid_3d === End of Compilation 4 === 1501-510 Compilation successful for file xsf.f90. mpixlf90 -o average.x \ average.o addusdens1d.o add_shift_cc.o add_shift_lc.o add_shift_us.o atomic_wfc_nc_proj.o cft.o cgracsc.o compute_ppsi.o compute_sigma_avg.o cube.o cubicspinsym.o dosg.o do_initial_state.o do_shift_ew.o d_matrix_nc.o d_matrix_so.o elf.o ggen1d.o hexspinsym.o local_dos.o local_dos_mag.o local_dos1d.o openfil_pp.o plot_io.o plot_whole_cell.o punch_plot.o smallgk.o start_postproc.o stm.o stop_pp.o sym_band.o work_function.o write_p_avg.o xsf.o ../Modules/atom.o ../Modules/basic_algebra_routines.o ../Modules/cell_base.o ../Modules/check_stop.o ../Modules/clocks.o ../Modules/constants.o ../Modules/constraints_module.o ../Modules/control_flags.o ../Modules/descriptors.o ../Modules/dspev_drv.o ../Modules/electrons_base.o ../Modules/error_handler.o ../Modules/exc_t.o ../Modules/fft_base.o ../Modules/fft_parallel.o ../Modules/fft_scalar.o ../Modules/fft_types.o ../Modules/functionals.o ../Modules/io_files.o ../Modules/io_global.o ../Modules/ions_base.o ../Modules/input_parameters.o ../Modules/kind.o ../Modules/mp_global.o ../Modules/mp_wave.o ../Modules/mp.o ../Modules/mp_base.o ../Modules/metagga.o ../Modules/parallel_include.o ../Modules/parameters.o ../Modules/parser.o ../Modules/paw_variables.o ../Modules/printout_base.o ../Modules/pseudo_types.o ../Modules/ptoolkit.o ../Modules/random_numbers.o ../Modules/radial_grids.o ../Modules/read_ncpp.o ../Modules/read_upf_v1.o ../Modules/read_upf_v2.o ../Modules/read_uspp.o ../Modules/recvec.o ../Modules/shmem_include.o ../Modules/stick_base.o ../Modules/splinelib.o ../Modules/task_groups.o ../Modules/timestep.o ../Modules/upf_to_internal.o ../Modules/upf.o ../Modules/uspp.o ../Modules/version.o ../Modules/vxc_t.o ../Modules/vxcgc.o ../Modules/wavefunctions.o ../Modules/wannier.o ../Modules/write_upf_v2.o ../Modules/xml_io_base.o ../Modules/zhpev_drv.o ../PW/libpw.a ../flib/ptools.a ../flib/flib.a ../clib/clib.a ../iotk/src/libiotk.a /bgl/local/lib/liblapack440.a /bgl/local/lib/libblas440.a /opt/ibmmath/essl/4.2/lib/libesslbg.a ../PW/libpw.a(stop_run.o): In function `stop_run': : undefined reference to `__path_io_routines_NMOD_io_path_stop' ../PW/libpw.a(stop_run.o): In function `stop_run': : undefined reference to `__metadyn_vars_NMOD_deallocate_metadyn_vars' ../PW/libpw.a(stop_run.o): In function `stop_run': : undefined reference to `__path_variables_NMOD_path_deallocation' make[1]: *** [average.x] Error 1 make[1]: Leaving directory `/gsa/yktgsa-h1/00/rajapand/espresso-4.0.5/PP' make: *** [pp] Error 2 I will greatly appreciate any help/suggestion. Thanks, Rajan K. Pandey IBM India SRDC, D3 First Floor, Nagawara, Outer Ring Road, Bangalore - 560045 INDIA -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090529/0bc66e72/attachment.htm From uccaati at ucl.ac.uk Fri May 29 17:34:44 2009 From: uccaati at ucl.ac.uk (Antonio Tilocca) Date: Fri, 29 May 2009 16:34:44 +0100 (BST) Subject: [Pw_forum] Saving the ion velocities Message-ID: Laura, as far as I know, a normal restart of a previous Nose-thermostatted MD run, with the thermostat turned off, does not reset the ion velocities to zero, unless you specifically request it in the input: if you send an input file/test case I could be able to tell more. Antonio -------------------- Antonio Tilocca Chemistry - UCL www.ucl.ac.uk/~uccaati ----------------------- > > Message: 1 > Date: Thu, 28 May 2009 16:20:24 -0400 (EDT) > From: Laura Henderson > Subject: [Pw_forum] Saving the ion velocities > To: pw_forum at pwscf.org > Message-ID: > <7110493.1243542024178.JavaMail.laura.henderson1 at mycampus.uoit.ca> > Content-Type: text/plain; charset=UTF-8 > > Hi to all, > > I have a question about molecular dynamics on Quantum Espresso. We use > the Nose thermostat to control the temperature of our system; however, > we also need to run free molecular dynamics without the Nose thermostat > on. When the thermostat is turned off, the ion velocities reset to > zero, so the system is quenched to zero Kelvin. Is there a way to turn > off the Nose thermostat without the ion velocities resetting to zero? > > Thanks > Laura > > --------- > Laura Henderson > University of Ontario Institute of Technology, Ontario, Canada > From giannozz at democritos.it Fri May 29 18:24:06 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Fri, 29 May 2009 18:24:06 +0200 Subject: [Pw_forum] Compilation Error on BlueGene/L In-Reply-To: <48b8698d0905290505y3fa5455dq6f58852003f5d1bb@mail.gmail.com> References: <48b8698d0905290505y3fa5455dq6f58852003f5d1bb@mail.gmail.com> Message-ID: On May 29, 2009, at 14:05 , Raj Pandey wrote: > I ran the configure script on BlueGene/L: > [...] > ../PW/libpw.a(stop_run.o): In function `stop_run': > : undefined reference to `__path_io_routines_NMOD_io_path_stop' > ../PW/libpw.a(stop_run.o): In function `stop_run': > : undefined reference to `__metadyn_vars_NMOD_deallocate_metadyn_vars' > ../PW/libpw.a(stop_run.o): In function `stop_run': > : undefined reference to `__path_variables_NMOD_path_deallocation' > [...] > I will greatly appreciate any help/suggestion. 0) compile only what you need 1) add missing modules. For the above case, add to MODULES in PP/ Makefile: ../Modules/path_io_routines.o ../Modules/metadyn_vars.o ../Modules/ path_variables.o P. --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From javier.camarillo at cimav.edu.mx Sat May 30 06:48:00 2009 From: javier.camarillo at cimav.edu.mx (Javier Camarillo Cisneros) Date: Fri, 29 May 2009 22:48:00 -0600 (MDT) Subject: [Pw_forum] atom occupied factor Message-ID: Hi to all, I have a question about atom occupied factor on Quantum Espresso. Is there a way to assign a number between .1 to 1 to two different atoms in the same position? Thanks Javier --------- Javier Camarillo Centro de Investigaci?n en Materiales Avanzados, Chihuahua, Mexico From javier.camarillo at cimav.edu.mx Sat May 30 07:33:05 2009 From: javier.camarillo at cimav.edu.mx (Javier Camarillo Cisneros) Date: Fri, 29 May 2009 23:33:05 -0600 (MDT) Subject: [Pw_forum] about atom occupied factor Message-ID: <52f611f58bd736ddc5aef8d553f012dd.squirrel@buzon.cimav.edu.mx> Hi to all, I have a question about atom occupied factor Is there a way to assign a number between .1 to 1 to two different atoms in the same position? Thanks Javier --------- Javier Camarillo Centro de Investigaci?n en Materiales Avanzados, Chihuahua, Mexico Javier Camarillo Cisneros CIMAV Student at M.C. Materiales (044614) 1520453 Chihuahua, MX From lanhaiping at gmail.com Sat May 30 10:06:35 2009 From: lanhaiping at gmail.com (lan haiping) Date: Sat, 30 May 2009 16:06:35 +0800 Subject: [Pw_forum] about atom occupied factor In-Reply-To: <52f611f58bd736ddc5aef8d553f012dd.squirrel@buzon.cimav.edu.mx> References: <52f611f58bd736ddc5aef8d553f012dd.squirrel@buzon.cimav.edu.mx> Message-ID: Do you just mean virtual crystal approximation ? If so, you can have a look at $QE/upftools/virtual.f90. But you should make sure some basic knowledge of virtual crystal approximation. regards, On Sat, May 30, 2009 at 1:33 PM, Javier Camarillo Cisneros < javier.camarillo at cimav.edu.mx> wrote: > > Hi to all, > > I have a question about atom occupied factor > > Is there a way to assign a number between .1 to 1 to two different atoms > in the same position? > > Thanks > Javier > > --------- > Javier Camarillo > Centro de Investigaci?n en Materiales Avanzados, Chihuahua, Mexico > > > Javier Camarillo Cisneros > CIMAV > Student at M.C. Materiales > (044614) 1520453 > Chihuahua, MX > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090530/ed00ba0c/attachment.htm From jwgonzal2 at yahoo.es Sat May 30 11:21:09 2009 From: jwgonzal2 at yahoo.es (=?iso-8859-1?Q?Jhon_W=2E_Gonz=E1lez?=) Date: Sat, 30 May 2009 09:21:09 +0000 (GMT) Subject: [Pw_forum] SCF correction problem in relaxation Message-ID: <515069.78539.qm@web27604.mail.ukl.yahoo.com> Hi, I'm trying to relax a cell to calculate transport properties of zigzag graphene nanoribbon, but there is something wrong, because for all parameters, always get the same final result SCF correction compared to forces is too large, reduce conv_thr This message appears for all conv_thr parameters (From 1.0D-6 to 1.0D-1) Thanks a lot, ########################## &CONTROL ... etot_conv_thr = 1.0D-4 , forc_conv_thr = 1.0D-3 , / &SYSTEM ... occupations = 'smearing' , degauss = 0.01 , smearing = 'fermi-dirac' , / &ELECTRONS electron_maxstep = 100, conv_thr = 1.0D-5 , / &IONS / ############################## Produces an output like: Total force = 0.001237 Total SCF correction = 0.000289 SCF correction compared to forces is too large, reduce conv_thr bfgs converged in 11 scf cycles and 8 bfgs steps End of BFGS Geometry Optimization ############################ &ELECTRONS diagonalization = 'cg', mixing_mode = 'plain', mixing_beta = 0.7, conv_thr = 1.D-6, mixing_beta = 0.3D0, / &IONS ion_dynamics='bfgs' bfgs_ndim = 1, pot_extrapolation = "second_order", wfc_extrapolation = "second_order", / ############################# Also produce the same warning SCF correction compared to forces is too large, reduce conv_thr Jhon W. Gonz?lez Universidad T?cnica Federico Santa Mar?a Fono (56) (32) 2654623 Fax (56) (32) 2797656 Casilla 110-VALPARAISO Avenida Espa?a 1680 239-0123 VALPARAISO CHILE for QE forum: http://www.democritos.it/mailman/listinfo/pw_forum From paulatto at sissa.it Sat May 30 12:29:56 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 30 May 2009 12:29:56 +0200 (CEST) Subject: [Pw_forum] SCF correction problem in relaxation In-Reply-To: <515069.78539.qm@web27604.mail.ukl.yahoo.com> References: <515069.78539.qm@web27604.mail.ukl.yahoo.com> Message-ID: <49976.78.12.170.167.1243679396.squirrel@webmail.sissa.it> Dear John, it means that your convergence threshold is too big: > conv_thr = 1.0D-5 , it is at least 10 and at most 10 thousand times larger than an acceptable convergence threshiold. Either increase it or increase the upscale parameter. regards P.S. you could also try and search the mailing list archive: please note that the same question has been asked and answered no more than 7 days ago. -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From anvari_meh at physics.iust.ac.ir Sat May 30 15:28:48 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Sat, 30 May 2009 16:58:48 +0330 Subject: [Pw_forum] VdM force Message-ID: Dear Paulatto Thanks a lot. I've studied a few about frozen-core approximation in the papers of Giannozzi about numerical methods in electronic structure. I think that this approximation is the first step to produce pseudopotential for a certain atom and then transfer it to particular crystal, but when I read your answer I became confused, because you wrote that the pseudopotentials & frozen-core are different types of approximation.(?) Best Regards Mehrnaz Anvari Iran University Of Science & Technology -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090530/3dc90796/attachment.htm From physolde at yahoo.fr Sat May 30 15:16:27 2009 From: physolde at yahoo.fr (kada galami) Date: Sat, 30 May 2009 13:16:27 +0000 (GMT) Subject: [Pw_forum] elastic properties of orthorhombic structure Message-ID: <117890.14222.qm@web26602.mail.ukl.yahoo.com> dear users last times when i ask about calculation of elastic properties of orthorhombic structure you answer me that we can do it by hand . I want to known how? for example : I have a monoclinic strain to apply to my structure and calculate total energy the question is which space group and atomic position i can use in file input thank you all -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090530/1dcf647c/attachment.htm From giannozz at democritos.it Sat May 30 15:29:33 2009 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sat, 30 May 2009 15:29:33 +0200 Subject: [Pw_forum] about atom occupied factor In-Reply-To: <52f611f58bd736ddc5aef8d553f012dd.squirrel@buzon.cimav.edu.mx> References: <52f611f58bd736ddc5aef8d553f012dd.squirrel@buzon.cimav.edu.mx> Message-ID: On May 30, 2009, at 7:33 , Javier Camarillo Cisneros wrote: > Is there a way to assign a number between .1 to 1 to two > different atoms in the same position? "a number between .1 to 1" meaning what? Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 From paulatto at sissa.it Sat May 30 17:16:00 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sat, 30 May 2009 17:16:00 +0200 (CEST) Subject: [Pw_forum] VdM force In-Reply-To: References: Message-ID: <51423.78.12.169.77.1243696560.squirrel@webmail.sissa.it> On Sat, May 30, 2009 15:28, Mehrnaz Anvari wrote: > because you wrote that the pseudopotentials > frozen-core are different types of approximation. Yes, the pseudopotential aproximation always (in practice) includes the frozen-core aproximation; but the opposite is not true. The PAW method formally falls in this category. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From aryjunior at gmail.com Sat May 30 20:49:38 2009 From: aryjunior at gmail.com (Ary Junior) Date: Sat, 30 May 2009 15:49:38 -0300 Subject: [Pw_forum] wrong record length In-Reply-To: <34ad5960903091603u12506823xc3f8c3a750d60002@mail.gmail.com> References: <34ad5960903020842h5f7edffncbf9d253d18035d9@mail.gmail.com> <34ad5960903021205j77f7a279j205cbfd6cbd906a1@mail.gmail.com> <49ACF6FB.3010207@democritos.it> <34ad5960903091603u12506823xc3f8c3a750d60002@mail.gmail.com> Message-ID: <4c9c61c20905301149o79ce34f6i81b3b3b77f3ff975@mail.gmail.com> Hi, I'm using ifort 11.0 64 bits to compile espresso-4.0.4 and I get the same "wrong record length"... Which is the equivalent "-qintsize=8" option for ifort compiler? Thank you very much! On Mon, Mar 9, 2009 at 8:03 PM, Marci wrote: > Dear Axel, Lex and Paolo, > > Sorry for the late answer. Thanks again for all your help. If someone > meets the same problem, here is the solution. Unfortunately the > easiest way didn't work (just outcommenting the test for negative > unf_recl) and declaring unf_recl as integer*8 didn't help either. > However on this specific system with XL fortran compiler adding > "-qintsize=8" to FFLAGS solved my problem. > > Yours, > Marton > > 2009/3/3 Paolo Giannozzi : > > Axel Kohlmeyer wrote: > > > >> the interesting question is now, how to work around this problem. > >> you could try and declare unf_recl as integer*8 and try to recompile. > > > > once upon a time there were limitations on the maximum length of > > direct access records on some machines. I remember I modified the > > two routines that open and read/write in such a way that each > > record was split into more records with the maximum allowed > > length. Anyway, for very big jobs, integers that don't fit > > into 32 bits will show up somewhere else, so the ultimate > > solution is to compile with default 64-bit integers, I guess > > > > Paolo > > -- > > Paolo Giannozzi, Democritos and University of Udine, Italy > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090530/dfae8d4e/attachment.htm From akohlmey at cmm.chem.upenn.edu Sat May 30 21:12:30 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 30 May 2009 15:12:30 -0400 Subject: [Pw_forum] wrong record length In-Reply-To: <4c9c61c20905301149o79ce34f6i81b3b3b77f3ff975@mail.gmail.com> References: <34ad5960903020842h5f7edffncbf9d253d18035d9@mail.gmail.com> <34ad5960903021205j77f7a279j205cbfd6cbd906a1@mail.gmail.com> <49ACF6FB.3010207@democritos.it> <34ad5960903091603u12506823xc3f8c3a750d60002@mail.gmail.com> <4c9c61c20905301149o79ce34f6i81b3b3b77f3ff975@mail.gmail.com> Message-ID: <1243710750.3912.3259.camel@zero> On Sat, 2009-05-30 at 15:49 -0300, Ary Junior wrote: > Hi, I'm using ifort 11.0 64 bits to compile espresso-4.0.4 and I get > the same "wrong record length"... Which is the equivalent > "-qintsize=8" option for ifort compiler? it is listed in the compiler help! -i8 or -integer-size 64 axel. > > Thank you very much! > > On Mon, Mar 9, 2009 at 8:03 PM, Marci wrote: > Dear Axel, Lex and Paolo, > > Sorry for the late answer. Thanks again for all your help. If > someone > meets the same problem, here is the solution. Unfortunately > the > easiest way didn't work (just outcommenting the test for > negative > unf_recl) and declaring unf_recl as integer*8 didn't help > either. > However on this specific system with XL fortran compiler > adding > "-qintsize=8" to FFLAGS solved my problem. > > Yours, > Marton > > 2009/3/3 Paolo Giannozzi : > > > Axel Kohlmeyer wrote: > > > >> the interesting question is now, how to work around this > problem. > >> you could try and declare unf_recl as integer*8 and try to > recompile. > > > > once upon a time there were limitations on the maximum > length of > > direct access records on some machines. I remember I > modified the > > two routines that open and read/write in such a way that > each > > record was split into more records with the maximum allowed > > length. Anyway, for very big jobs, integers that don't fit > > into 32 bits will show up somewhere else, so the ultimate > > solution is to compile with default 64-bit integers, I guess > > > > Paolo > > -- > > Paolo Giannozzi, Democritos and University of Udine, Italy > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From anvari_meh at physics.iust.ac.ir Sun May 31 04:52:40 2009 From: anvari_meh at physics.iust.ac.ir (Mehrnaz Anvari) Date: Sun, 31 May 2009 06:22:40 +0330 Subject: [Pw_forum] choosing ecutwfc through vc-rx Message-ID: Dear all I have some fundamental questions. 1.If I want to achieve the equilibirium value of lattice paramrters (such as a, c,...) is it possible to replace bulk modulues calculations with vc-rx? 2.when I use vc-rx for my material most of the time after 300 stpes the stress becomes approximately zero(-0.6-0.6) but energy doesn't converge. I don't pay attention just use the cell parameters & atomic positions to do scf calculations. I tried to achieve convergency in scf calculation with changing ecutwf & k-points. Now I think it couldn't be true & I have to optimize my ecutwfc during vc-rx calculation (Is it true?) what about k-point ,where could I optimize it? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090531/79d21b11/attachment.htm From paulatto at sissa.it Sun May 31 14:17:57 2009 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Sun, 31 May 2009 14:17:57 +0200 (CEST) Subject: [Pw_forum] choosing ecutwfc through vc-rx In-Reply-To: References: Message-ID: <57873.78.12.159.5.1243772277.squirrel@webmail.sissa.it> On Sun, May 31, 2009 04:52, Mehrnaz Anvari wrote: > 1.If I want to achieve the equilibirium value of lattice paramrters > (such as a, c,...) is it possible to replace bulk modulues calculations > with vc-rx? Yes, that's what it's made for. Anyway, it may be a good idea to compute the bulk modulus anyway, in order to test your setupe and your pseudopotentials. > 2.when I use vc-rx for my material most of the time after 300 stpes the > stress becomes approximately zero(-0.6-0.6) but energy doesn't converge. There is some error in you input, or you have bad pseudopopotential. Probably you are using damped-dynamics that are not damped at all. > Now I think it couldn't be true & I > have to optimize my ecutwfc during vc-rx calculation (Is it true?) what > about k-point ,where could I optimize it? Just take a reasonable configuration to test convergence and than relax to fine-tune the parameters. regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From sophia.nishad at gmail.com Sun May 31 16:27:02 2009 From: sophia.nishad at gmail.com (Sophia Nishad) Date: Sun, 31 May 2009 07:27:02 -0700 Subject: [Pw_forum] gamma phonongs for alloys Message-ID: <8a0137e60905310727y59b0257ajb42b6e37f633512f@mail.gmail.com> Hi, I wanted to know what is the best approch to compute gamma phonons for alloys with the linear response thoery. To include disorder effects how large the supercell has to be for a model alloy say, SiGe. Would the result converge for a supercell of 64 atoms? or more atoms are needed. Would it be efficient to use mass approximation or any other tricks like VCA approximation for the pseudopotential? Also, how about any related properties like strain-shift coefficents. Would they be sensitive to size of supercell as well? Thanks in advance, -- Sophia -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090531/989630af/attachment.htm From akohlmey at cmm.chem.upenn.edu Sun May 31 19:10:07 2009 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 31 May 2009 13:10:07 -0400 Subject: [Pw_forum] choosing ecutwfc through vc-rx In-Reply-To: References: Message-ID: <1243789807.3912.4593.camel@zero> On Sun, 2009-05-31 at 06:22 +0330, Mehrnaz Anvari wrote: > Dear all > I have some fundamental questions. > 1.If I want to achieve the equilibirium value of lattice paramrters > (such as a, c,...) is it possible to replace bulk modulues > calculations with vc-rx? in principle, yes. but there are some caveats. for vc-relax you have to converge the stress tensor and that will usually take a higher wavefunction cutoff and if you use ultrasoft psps particularly a higher density cutoff. so for all simple systems where the symmetry defines all/most atom positions, just running a series of single point total energy calculations should be much less computational effort and is trivially parallizable. > 2.when I use vc-rx for my material most of the time after 300 stpes > the stress becomes approximately zero(-0.6-0.6) but energy doesn't > converge. I don't pay attention just use the cell parameters & atomic > positions to do scf calculations. I tried to achieve convergency in > scf calculation with changing ecutwf & k-points. Now I think it > couldn't be true & I have to optimize my ecutwfc during vc-rx > calculation (Is it true?) what about k-point ,where could I optimize > it? > wavefunction cutoff and k-point grid can be converged fairly independently. however, it is usually more efficient to first converge the wavefunction (and density) cutoff with just one or a few k-points and then crank up the k-points and make a confirmation test on wavefunction. if you use smearing, then you also have to consider the smearing parameter vs. number of k-points relation. cheers, axel. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From javier.camarillo at cimav.edu.mx Sun May 31 20:35:55 2009 From: javier.camarillo at cimav.edu.mx (Javier Camarillo Cisneros) Date: Sun, 31 May 2009 12:35:55 -0600 (MDT) Subject: [Pw_forum] about atom occupied factor Message-ID: <2687ee6de6958ea3017f94505a0e7359.squirrel@buzon.cimav.edu.mx> Hi everybody I am sorry for my brief last message The occupied factor between .1 to 1 is the factor of times that a atom appears on one position, shared with other atom. For example, I need specified two different atom (Fe 25% and Ti 75%) to the same position at one octahedral site on Aurivillius crystal. This crystal is formed to 196 atoms, so is not possible create 4 times the lattice to incorporate at the same time the 3 Ti and the 1 Fe. Thanks to the last 3 answer to the last message: Mrs Ian, Paolo and Stefano > Hi to all, > > I have a question about atom occupied factor > > Is there a way to assign a number between .1 to 1 to two different atoms > in the same position? > > Thanks > Javier > > --------- > Javier Camarillo > Centro de Investigaci?n en Materiales Avanzados, Chihuahua, M?xico > > From: lan haiping Do you just mean virtual crystal approximation ? If so, you can have a look at $QE/upftools/virtual.f90. But you should make sure some basic knowledge of virtual crystal approximation. From: Paolo Giannozzi "a number between .1 to 1" meaning what? From pushpajnc at gmail.com Sun May 31 22:52:03 2009 From: pushpajnc at gmail.com (pushpa raghani) Date: Sun, 31 May 2009 13:52:03 -0700 Subject: [Pw_forum] Is 'stm_wfc_matching' option operational? Message-ID: Hi, Is the option 'stm_wfc_matching' operational in post processing? In 3.2.3 version, it says that stm_wfc_matching is presently disabled. In 4.0 version, it doesn't say anything, so I guess it should be working there? However, I get the error 'option stm_wfc_matching does not work'. I tried this both on symmetric and asymmetric slab. Any ideas? Many Thanks, Pushpa (Stanford University) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090531/cecdc5f6/attachment.htm From duchl06 at 163.com Sun May 31 23:18:26 2009 From: duchl06 at 163.com (duchl06) Date: Mon, 1 Jun 2009 05:18:26 +0800 (CST) Subject: [Pw_forum] will mixing_beta affect the calculation result? In-Reply-To: References: Message-ID: <9758779.11461243804706825.JavaMail.coremail@bj163app41.163.com> Dear pwscf users i thought the mixing_beta only affect the convergence velocity, but when i did a scf calculation recently, i got different result using different mixing_beta (the total energy have a disparity about 0.03ry when the mixing_beta chosen 0.7 and 0.2, this is very huge ), i do not know what happened. will mixing_beta affect the calculation result? are there any error in my calculation or i have a misunderstanding on the mixing_beta? thank you! Du -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20090601/e57ae53e/attachment.htm From bogus@does.not.exist.com Wed May 27 16:26:59 2009 From: bogus@does.not.exist.com () Date: Wed, 27 May 2009 14:26:59 -0000 Subject: No subject Message-ID: call elphsum() if(ionode) then call check_routine() endif call mp_barrier() . . . From bogus@does.not.exist.com Wed May 27 16:26:59 2009 From: bogus@does.not.exist.com () Date: Wed, 27 May 2009 14:26:59 -0000 Subject: No subject Message-ID: call elphsum() if(ionode) then call check_routine() endif call mp_barrier() . . . From bogus@does.not.exist.com Wed May 27 16:26:59 2009 From: bogus@does.not.exist.com () Date: Wed, 27 May 2009 14:26:59 -0000 Subject: No subject Message-ID: Molecular Dynamics', while the 'md' calculation in pw.x performs born-oppenheimer MD. Paolo just told us that the born-oppenheimer MD only perform MD at ground state. Does CPMD or cp.x perform the MD calculation at ground state or exited state? thank you very much for you time best, vega -- ================================================================================== Vega Lew ( weijia liu) PH.D Candidate in Chemical Engineering State Key Laboratory of Materials-oriented Chemical Engineering College of Chemistry and Chemical Engineering Nanjing University of Technology, 210009, Nanjing, Jiangsu, China ****************************************************************************************************************** Email: vegalew at gmail.com Office: Room A705, Technical Innovation Building, Xinmofan Road 5#, Nanjing, Jiangsu, China ****************************************************************************************************************** --0016362837525fd150046cfd59f0 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Dear friends,

Thank you for you=A0 time for reading and answering my= questions.

I realized that the ensemble is much more important than= one snap of the whole run, resembling the classical MD to some extent.

I used to ask "if I could achieve the same final geometry for same= systems (the coordinates of atoms, temparature, box dimensions and K-ponits etc. are exactly the same) by this two methods."

I didn't mean the s= ame coordinates of the final systems. I wonder if the different methods wit= h same input coordinates will give us the same conclusion. For example, By = relax calculation by pw.x water disociative adsorption is extremely favorab= le at certain metal surface with a minute barrier of ~0.20 eV given by NEB = method. But before the calculation we didn't know this. Suppose we know= nothing about it before the ab initio MD calcualtions. We only take ab ini= tio MD, and we put a integrated water molecule at certain site on the surfa= ce. and then we perform MD calculations of different method, such as CPMD, = cp.x, pw.x(calcualtion=3D'md'). I wonder whether the dissociative s= tate could be acheived on the surface at very low temparature (such as 100K= or below) via all the MD calculations mentioned above?

From your answers, I see cp.x and CPMD of IBM CORP. is using 'Car-P= arrinello Molecular Dynamics', while the 'md' calculation in pw= .x performs born-oppenheimer MD. Paolo just told us that the born-oppenheim= er MD only perform MD at ground state. Does CPMD or cp.x perform the MD cal= culation at ground state or exited state?


thank you very much for you time

best,
vega


--
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Vega Lew ( weijia liu)<= br>PH.D Candidate in Chemical Engineering
State Key Laboratory of Materials-oriented Chemical Engineering
College of Chemistry and Chemical Engineering
Nanjing University of Tech= nology, 210009, Nanjing, Jiangsu, China
********************************= ***************************************************************************= *******
Email: vegalew at gmail= .com
Office: Room A705, Technical Innovation Building, Xinmofan Road= 5#, Nanjing, Jiangsu, China =A0
***************************************= ***************************************************************************=
--0016362837525fd150046cfd59f0--